“After seeing the results of these tests, we can only conclude that Apple’s fastest G4 workstations are certainly not faster than dual 1.533MHz Athlon MP-equipped systems, at least as far as After Effects is concerned. Not one of the objective tests we conducted using After Effects showed Apple’s flagship machine to be superior. In fact, in most of the tests, the Mac was left lagging far behind. And, this was when we were using a PC whose chips are now three steps below the fastest Athlons available (now the fastest Athlon XP is the 2100+ running at 1.73GHz), and far slower than benchmarks have shown the latest Intel Xeon chips to be, now shipping at 2.4 GHz.” The benchmark is held at DigitalVideoEditing. Our Take: We would like to see the benchmarks with dual Athlon MPs or dual PIIIs at 1 Ghz, so it would be more fair, Mhz-wise. On the other hand, Mhz doesn’t matter. Right Steve? And at the end of the day, this was the fastest G4 available. The Dual AthlonMP 1.5 is not the fastest available anymore in the PC world (dual SMT Xeon 2.4 Ghz at the same price as the dual G4 1 Ghz anyone?). BTW, it seems that the Mac users feel left behind these days.
Dual AthlonMP 1.5 Ghz Significantly Outperforms Dual G4 1 Ghz
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
get Motorola off of their butt and get the G5 out in production.
Well, according to The Register, G5s were able to run from 1.2 Ghz up to 1.6 Ghz in the Motorola factory and its first generation had problems making the G5 run on a faster clock. By the time G5 1.6 Ghz comes out, Intel will be selling 3 Ghz with even faster bus speeds, and once again, it will kick butt.
To be clear: G5 is good. Problem is, it is too late already, Apple & Motorola haven’t even dared to announce anything, possibly because it is not ready yet. By the time is out, it will also be outperformed by Hammer or McKinley or Xeons or P4s at 3 Ghz.
And that is the real problem, not the fact that G5 is good or bad. And you can’t just say: “G5 will show you!!”, because there are no announcements for it whatsoever. It is like talking about vaporware, EVEN if it is indeed exists. Intel and AMD have already prototypes and officially published roadmaps about their upcoming chips and their speeds.
…Apple’s fastest G4 workstations are certainly not faster than dual 1.533MHz Athlon MP-equipped systems…<BR><BR>Either the G4s perform abismally or those Athlons are incredibly efficient. They’d have to be capable of around 1,000 operations per cycle. How much are they going for? Do they also offer higher clock speeds?
are stupid. I can do just fine on a machine runing at 800MHz as I would be able to on a machine runing 5 GHz.
E-mail…Check
Web browsing….Check
Games….Check
GUI…Check
Word Processing and office productivity……Check
Programming…..Check
Music/MP3s….check
DVD Movies….check
all run just fine on a 1GHz machine, no matter the proc. why do I need to spend big bucks on a proc that will not give me any performence upgrade in actual use?
I mean yeah, in the days of the 486/P1/P2 new procs could improve your computer substatialy, but now, the hardware has so outpaced the software, that the lowest end machines can run the software with easy.
I thought Apple’s 1Ghz G4 claim was that it was as fast or faster than a 1.5Ghz machine? Am I wrong? What was the machine speed that Jobs compared the G4 to at his infamous talk last year?
Anyways, sure 1.5Ghz is not “really” a fair comparison to 1Ghz, but it isn’t “really” fair to compare a PPC to an Intel chip– different processes completely.
What this does prove, in my eyes, is that Apple really needs to do some serious re-tooling in their thought process and either get their Ghz up or get a new chip.
Paul
is it me or does someone really hate apple? There is a constant influx of these “apple performance is a lie” type of articles.
I encourage the editors (who have now done the apple sucks thing to death)to consider articles looking at something else.
> all run just fine on a 1GHz machine, no matter the proc. why do I need to spend big bucks on a proc that will not give me any performence upgrade in actual use?
Jeremy, please stop the cheap excuses. This web site is the DigitalVideoEditing. TIME costs for a video or gfx professional. Exactly why you spend more money to buy better development tools that happen to be faster. Because time is money for professionals, and they need the fastest possible.
Sure, I got the G4 Cube at 450 Mhz. It works fine for my personal needs as a desktop user. But it doesn’t cut it as a server, it doesn’t even cut it as a geek machine anymore if I want to run something more demanding, I am limited by its speed.
We have discussed with JBQ that our next PC here at home will be a dual Xeon SMT. It is CHEAP for what it delivers. My dual Celeron 533 is great for all I do these days, but try running VMWare or a server with it, it feels old…
Things change, they evolve. Problem is, Apple evolves much less hardware-wise these days. And it is a shame really.
>I encourage the editors (who have now done the apple sucks thing to death)to consider articles looking at something else.
Put a sock in it.
We already discussed this. There are articles that are against and FOR Apple. They will always be. As they always are for the other platforms too.
Please consult the archived messages for this discussion: “OSnews is against Aplpe” “OSNews is not against Apple”
It is getting boring.
I’d be curious to see a power consumption comparison as well. My guess is that the G4’s consume considerably less power than the Athlons for the same amount of processing.
Eugenia’s 450 MHz G4 doesn’t even have a fan, correct?
Personally, I don’t like my personal computer to double as a space heater.
I’m a Mac user. My main (and fastest) machine is an iBook500. Guess what? It’s good enough for me. I’m a developer, and I use Squeak for the majority of my work. Works good enough for that. I know I could get a machine that runs Squeak faster, but it’s not a big deal. Because I still have to deal with the rest of the computer, and the host OS. My time is worth more than to have to put up with Windows and Linux and all the inconsistencies and instability (of Windows). I just want to get stuff done, not jibberjabber about having the highest MHz machine.
Maybe if BeOS wasn’t dead (or getting there), I would’ve stuck with it on x86 hardware. But before I switched to Mac OS X, I was using Linux. And it sucked. Still does, when I need to get ‘normal’ desktop stuff done. I’m a big geek, and *can* do everything I need to under Linux, but there’s a lot of time overhead. Just because I can, it doesn’t mean I want to deal with the pain in the ass and time it takes to get there.
I may switch back to x86 as soon as I can get Squeak doing everything I need it too, so I don’t have to screw around with a sucky host-OS. But even then, I’d be embarassed to run such an inefficient machine, as is the current x86 designs. The Oqo looks good as my next computer, and is a lot closer to being acceptable.
Maybe I’m a rare breed, a pragmatist. I think of it like a car- I’d much rather buy one that worked just like I wanted it to, didn’t need maintenance very often, got good gas milleage. I certainly wouldn’t base my descision in buying a car on how fast it went, as long as it went fast enough for the roads on which I drive.
It’s great that the kids like to have penis contests over how fast their machines are, but it’s just not my hobby.
I suppose I can see wanting a super-fast machine if you’re coding in a gimpy language like C, C++ or Java, where you have to make entire classes, files or trees for small changes in code. I’ve developed C++ on my 500 MHz iBook and on a 1.3 GHz Athlon. G++ is a lot quicker on the Athlon. But fortunately for me, Smalltalk doesn’t have to do that, each method is compiled when you save it in a fraction of a second, such that one doesn’t really notice the compilation step. /me hugs incremental compilation…
Jeremy: Processor Pi$$ing Contests are stupid. I can do just fine on a machine runing at 800MHz as I would be able to on a machine runing 5 GHz.
Well, I for one am tired of my full builds taking 6 hours on my PIII 800/512. Even builds for just my project take over 30min to do full debug and ship, and that is assuming you aren’t using the machine to do *anything* else.
Also, have you tried to play Dungeon Siege on a PIII 500 Xeon? In the case of this particular machine, the CPU is pegged all the time. It isn’t a grapic card issue either folks, I get 9fps in 1024×768, or 11fps in 800×600… either is abismal, and I know my GForce can do better than that.
Give me a 2.4GHZ Xeon any day.
I suffer from proc compairson fatuige. that is all.
I just hate the crap. probobly because most of the comparisons are done from a normal end user POV. and it is just pointless as I point out in my previouse post since all normal user apps run fin at the low end.
for a graphics application, fine I can accept the comparison as having a point, but other comparisons are used more by the idiot campers, who think <insert proc here> is better than the rest, to put gain pride points over the other campers.
it is just a stupid flamefest that for the most part has no point any longer, though this article is part of the minority that still does have a point.
I was just lashing out
sorry.
“It is getting boring.”
I think what is getting boring is the constant apple bashing. Bash someone new there are plenty of candidates.
The Athlon’s EV6-based system bus is superior, and could in fact be even better! The upcoming 64bit AMD Hammer will support NUMA (Non-Uniform Memory Access), which is even more advanced than EV6.
I saw linux running on a brand new athlon machine and my mind was completely blown, I’ve never seen a computer with I/O that fast.
the Xeon is not optimized for gaming.
try it on an Athlon 500. oh and it could verywell be your Video card. is it an origional GF? how old is the game?
Rambus memory also has high latency when compaired to SDRAM.
and if your Xeon is using SDRAM, it has a very Slow Bus Speed as 100 MHz was the fasted bus at the time (perhaps 133, but anyway)
Good to see someone else who expects more from their
computer! I hate resource waste. Call me crazy. Rather,
maybe I’m just plain un-American that I don’t try to waste
resources that aren’t really needed to bring me to my
desired end. Big-ass-hog computers and huge CRTs use up
an enourmous amount of energy use up as much as a fridge, or some high-fraction of the same amount of energy. Damn. That’s unacceptable. I can see if you are some silly American glutton that likes to use as much as you can, just because it’s there, but when there are other options, more efficient and elegant options, I go for them.
Back when PC users were totally dependent on Intel for CPUs and chipsets, it was easy for Apple to maintain a performance lead with the efficient PowerPC RISC architecture. Intel was not worried about Apple/IBM/Motorola, since they dominated 90%+ of the market.
However, around the time the Pentium I came out, AMD began challenging Intel with compatible chips, either offering faster execution or lower price. So the “power struggle” began. Pentium, K6-2, Pentium II, K6-3, Pentium 3, Athlon, Pentium 4, etc… Intel doesn’t want to be undercut and lose marketshare, or have their “image” damaged by faster processors from a small company like AMD.
Meanwhile, Apple is happily pulling along Mac users in a shiny red toy wagon, telling them that the G3 (et al) eats Pentiums for lunch… Too bad they hadn’t heard of the Penitum III, or Athlon, or even the K6-2 for that matter…
Apple has been lagging in performace for a long time, because it doesn’t see any direct competitors. Apple’s marketshare is based on brand loyalty (zealotry), fashion, and general uniqueness. They are too far behind now performance-wise, to ever catch up. Neither Motorola nor IBM have enough interest continue to put large amounts of R&D into the PPC architecture, only to make a relatively small number of chips for a niche market. It ain’t gonna happen.
If processor design merit was important to the industry (or consumers) at all, we should all be running Alphas; without dispute, the best/fastest/coolest processor architecture ever developed. Now R.I.P. eternally thanks to HP + Compaq, who realize that making the best CPU in the world isn’t worth it for such a small market.
And don’t give me that RISC vs. CISC irrelevancy… AMD’s chips have been using a very efficient RISC core since the K6-2. They had floating-point SIMD (yes, vector) since the K6-2, long before “AltiVec”. For Intel, the PIII added this (SSE), and of course the P4 has a high-speed RISC core as well. With large, full-core-speed on-chip caches, the x86 decode logic overhead is almost nil; the chips operates and perform just like highly parallel RISC devices.
Yes, I realize that the G3/G4 still makes good use of it’s 100MHz bus/RAM, etc, but its days are numbered. On my Athlon 1.4 (KT266A), I can pull over 1GB/sec from system RAM. That is well over 10x the performace you’re likely to squeeze from existing PC100 designs.
And >Jeremy<… you think this doesn’t matter? You’re living in the past. Sure, I can check my e-mail on a 386SX if I want. But this perpetual performace increase DOES make a difference for real-world applications, especially for power users. Anything that you have to “wait” for, gets that much faster… Photoshop filters, Video FX Rendering, 3D modeling, Game FPS, DivX encoding… and the applications will increase. With widespread broadband, I won’t be surprised to see broadcast-quality video-conferencing, and you won’t be doing that on a 800MHz box, probably not even a 1GHz box. And you just *might* want to be doing something else at the same time, a game, DVD, etc. or whatever… And I know your OS X GUI could sure use a faster CPU for the time being!
And >Chris<… what are you smoking?
1000 operations per cycle? Let me borrow your multi-million dollar supercomputer sometime, I’ve got some POV I’d like to run on it. Does IBM know about this? How about the US government?
Seriously, current Athlon and P4 designs are in the 1-3 GHz range, with bus speeds from 200MHz – 566 MHz. I’m not a real expert on the P4, but I know that the Athlon can perform up to 9 operations per clock cycle. At 1.5 GHz, that’s 13.5 BILLION instructions per second. 18 BILLION 16-bit multiplies per sec (MMX). 9 BILLION 32-bit floating-point multiplies per sec (3DNow!). Of course these are peak limits, but even unoptimized software should be able to exploit half of that, which is astounding.
PPC was nice while it lasted. Perhaps Apple should consider a port to AMD’s new Opterton… hey, that’s not such a bad idea… since it won’t be “Intel Compatible” (in the perpetual sense) it might go over better…
Actually, I’ve read up on a few of the news groups, and one comparison was made between a P4 1.4 Ghz with a GeForce 4Ti and a P4 2.2 Ghz with a GeForce 2 [something]. The 2.2 was outperforming the 1.4 by a margin of about 30%.
There’s just something about the game that is really CPU intensive. Maybe it is the auto loading of terrain or something.
Users will *always* require more power than they currently do. Sure e-mail and web browsing doesn’t require that much. How about realtime playback of DV quality video, or other tasks. How about all the high end performance applications which still don’t have enough horsepower, regardless of the computer it is on.
Bottom line is that Apple needs to make their scores for their top end machines *faster* than the fastest PC. Now that they have a decent operating system, they have an opportunity to get people to switch, if they also provide the added speed boost to go along with it. Right now they don’t. I’ve said this ad nauseum. As a Mac user I think they need to do something drastic within the next year to avoid getting left totally behind. Steve and Co. know this too, and I’m sure they are taking care of it. They just may not let us know quite yet.
It makes it too much sound as if this was a processor vs processor test, while it’s also an MacOS X vs Windows XP test. The fact is that MacOS X is a real performance killer, so IMO that’s a very important factor to take into account as well.
<snip>
E-mail…Check
Web browsing….Check
Games….Check
GUI…Check
Word Processing and office productivity……Check
Programming…..Check
Music/MP3s….check
DVD Movies….check
all run just fine on a 1GHz machine, no matter the proc. why do I need to spend big bucks on a proc that will not give me any performence upgrade in actual use?
</snip>
Granted for the most part the average/typical user you are correct. But there are people out there who like to:
Check email
Surf the Web
Listen/encode mp3’s
Encode a movie
Watch a DVD
Burn a CD’s
Program
etc…
in some combination at the same time. Duallies with as much power as you can get, gets the “job” done and it does it smoother.
One thing nobody seems to notice or count is the bus speed.
Any video app requires going to memory a lot and the speed you can do this at makes a difference.
This is one area where Apple / Motorola lags badly.
The G4s all stil have a 133MHz bus where as Athlons use 266MHz DDR which actually runs at the same speed but transfers at twice the rate, I suspect this is one of (if not the) major reason the Mac did so bad in this test.
There are PowerPC chips with RapidIO and 333MHz DDR support already on the way or in production. The G5 may be vapor so far but the 8540 is not:
http://e-www.motorola.com/webapp/sps/site/prod_summary.jsp?code=MPC…
We should see the same technologies appear in the 74xx series (which Apple uses) soon. The faster OS and compiler combination in OS X 10.2 should also help out considerably.
I didn’t mean to be unecessarily unpleasant to our host. It is just that these tests obscure a larger reality. Processor speeds have progressed to such a point that the p4 vs G4 thing doesn’t matter much. Today, it is really about offering a solution. There is a lot more to a solution than just the processor.
Fellow Mac users…we gotta stop kidding ourselves. No one [should] buy a Mac because they’re under the impression that it’s fast. Rather, I think there is a much better solution.
Get a cheaper Mac than you normally would, and spend the difference on an AMD/Intel box with a high-speed chip and RAM setup, and maybe a respectable GFX card. We all love the Mac interface, so get all your stuff modeled/edited/whatever on your Mac, then use the PC to render over a network. With Gigabit Ethernet, the overhead is minimal. I do this with Bryce all the time, and my PC is 2-3x faster than my G4/800.
So, let’s look at the economics of this:
G4/933 (CD-RW, Radeon 7500): $2000
Dual AthlonMP: $1500
Vs.
Dual G4/1000: $3500
Same price…and you get a lot more done. And you can play PC games.
Yeah I actually ran across this on ‘XLR8yourMac’ the other day and was going to give the link to ya Eugenia, but I thought the story was sort of old from being near the botom of their website. Yeah those numbers woke me up alittle, but I wonder if the CPU is the problem or the system bus and memory. After reading EETimes articles on the strength’s and weaknesses on the G4, it seems that the system bus these days is holding back its full potential because of the PCI tech has been a bad design since the beginning (EETimes opinion anyways). There has been a rumor that Apple is getting DDR fully on-board (not just the L3 backside cache). Here is a link of something that was about to sell on eBay until Apple pulled the plug…
http://www.macrumors.com/pages/2002/05/20020502153139.shtml
This was all over the tech sites that I usually wonder thru. So we hope this is a good sign.
I thought to myself after reading this article if I would just switch back to the PC platform just for CPU performance alone?… I still came to the conclusion of ‘NO’!
Me Mac Happy
🙂
Read the Apple newsgroups and the Apple message boards on web sites, and such news always produces two sets of apologetics:
1. Its a MHz myth propagated by the Intel/Microsoft machine
2. Who needs all this power anyway
Do most Mac users really have their head that much up their own arses that they don’t see the issues presented to them. The deciding factor whenever someone buys anything is how much bang they get for the buck. A person laying out $1500 will compare the MHz first and say that the Macs are far behind. This may be unfair from a hardware performance standpoint. However, now the specs for the $1500 Mac vs $1500 PC show that the PC user is getting more hardware processing power for their buck. Regardless of if a given G4 at the same MHz does more than the same P4 or Athlon, the fact is the systems per dollar are weaker on the Mac side. Couple this with the fact that the next generation graphics system on OS X requires more power, and that means that the Mac needs to have more powerful software to keep up to speed with the XP laden Athlon or P4.
To this, now the Mac users start throwing idiotic statements like “who needs that much power” or “how much energy is that P4 using versus the G4”. First problem with those arguments is that there is no such thing as too much computing power, memory or storage space. If you have more than you need, wait five years, you’re demand will soon catch up to your supply. Not too many people, especially those buying systems over $3000, fall into the category of having too much power. Therefore the raw performance, not even cost, is their driving purchasing variable. They will be woefully disappointed that the top end Mac, regardless of cost, doesn’t match the performance of the top end PC. As far as the power requirements of the chips, another big who cares. If I was so concerned about power consumption, I’d just fire up my Apple II’s 6502 and be happy with a fraction of a Watt of power, given todays manufacturing processes. For laptops, the battery life would be a possible deciding factor on this. However, if that laptop is only capable of giving me half the processing power, then its once again another big who cares.
The reason why I am so displeased about this announcement has nothing to do with a company conspiracy by Intel, or because I have some deep loathing for Intel chips. It comes from the fact that for the first time in 15 years, Apple is gaining momentum. They are gaining momentum on the OS front and on the device front. They are furthermore getting development houses which abandoned the Mac back into the fold. This momentum will only exist for a short time. In that short time, Apple has failed to produce enough iMacs to meet demands and it is not upgrading their hardware fast enough.
The bottom line is this, by this time next year, if the raw performance numbers of the Macintosh are not at least neck-in-neck in *all* operations (not just a handful of Photoshop filters) and if they don’t have a viable long term hardware development timeline, then this window of opportunity will evaporate. At that point, we will once again be left with a beleagured<sp> company with a short time to live.
What Apple needs, is a G6. As Eugenia pointed out, the G5 will be too little too late…
First of all let me say that I respect you a lot Eugenia, especially for what you’ve done for the BeOS community.
However:
a) Linking to a Macnn thread topic is another low for this site… These forums are notorious for the 14 year old whiners having endless and pointless dicussions on irrelevant subjects like the naming scheme for OSX, etc…
b) There is no equivalent for Final Cut Pro on that pc. There is no equivalent for DVD Studio Pro on that pc. The software that even comes close to offering comparable features are double the price. People do not buy Macs for the MHz, but for the OSX, and the software.
c) Do show me where you can configure a pc with dual 2,4GHz Xeons… I tried Dell and I’m afraid to tell you it came up to 5335$, against a 2999$ PowerMac. Similarly specced, of course, dvd-rw, gigabit ethernet, i put a Radeon VE in there because they had no GF4MX, no monitor.
d) you quote Steve’s “MHz myth”. Not a single mac user believes a dual GHz G4 to compete with those machines, except on specific tasks such as the fabled PS filters and mpeg2 encoding (which this site fails to mention, that Mac does it at over 1x speed…
We *know* those machines spank Macs at most tasks. The tone of your article makes it sound like a bad thread from the Ars Battlefront. The point of having a Mac is elsewhere. And the Video industry has understood that pretty well.
I suspect the DV guys messed up the benchmarking but good. I haven’t seen a PC beat a Mac in a proper speed test in 5-6 years. Having said that, I hope MOT gets off its broad and soft BUTT and gets the G5 out the door; Wintel does not even HAVE a credible 64 bit chip at this time. MOT is frustrating partner because they don’t even seem to comprehend what their main business is. Here is a chip collossus that is stuck in the mode that $50 cell phones are its mission in life. Morons!
quote: “it seems that the Mac users feel left behind”
I do. In fact a lot of Apple users do. But I’ll stick with Apple.
1st off, I use a mac. Its not a powerhouse (g3 400mhz 384MB RAM). I like it. I like OS 9. I *really* like OS X.
Is there faster? YES.
Is their something better? NO. Not for my needs and taste. Thats all the reasoning I need.
These Mac people crying out that Mot/IBM chips use less power than Intel/AMD chips . . .
Man are they THAT desperate to say ‘mines better than yours’. Who gives a rats@ss.
Fine if you are that concerned with energy efficiencey go by a VIA C3 😉
THATs the benchmark I want to see: VIA C3 versus G4. Then MAYBE the Mac freaks will have something to brag about. . . “When an x86 chip mathes a G3/4 in power efficiency, its overall performance is reduced below the g3/4, as PROVEN in the APPLE v Cyrix benchmarks . . . “
I would say IMHO that MIPS is the best processor ever designed so far.
I wonder how computing would be now if MIPS not x86 had won the battle.
For example there is a new MIPS processor out now that is a dual core, 64 bit, and consumes 7 yes thats seven watts of power at 1GHz.
At 600MHz it uses 1.5 watts of power.
Now imagin taking this processor and scaling it up for desktop use.
This MIPS processor makes PPC look pretty bad.
http://industry.java.sun.com/javanews/stories/story2/0,1072,42678,0…
I know you will understand this, being a mac user and all.
the Mac’s total experience is the selling point for it, if they ever move over to that new AMB proc that is not x86 or if IBM . MOTO gety a G5 out the door, the increased power will just be some extra goodness to be happy about.
realy, I mean it when I say I do not need all that power, or even care to spend the cash on it.
I have a 900 MHz Duron PC running Debian. I love it, it runs fast and does everything in the time I need it to.
hell, I had just upgraded the mobo and proc from a K6-3 400 that I bought brand new, so that tells you how long I had that sucker (I realized I needed an upgrade when XP would not load on my partition). I even squeez the last bit out of my Video. I did not upgrade my Video card from a Monster Fusion AGP 1x running the banshee chipset until I tried to play Black and White and the graphics choked.
I love my 400 MHz G4 Tower, I plan to upgrad the video in it so as to take full advantage of Jaguar, and the programs load perfectly fine.
Well it seems to be one of those back and forth issues… IBM just got a big pat on the back for the effort on the Power4 series (which is based on the PowerPC techno) and it beat both Intel’s Itanium and Compaq’s Alpha technology in performance marks. So it isn’t that I don’t have faith in the PowerPC technology, it is, but Motorola (or IBM) have to decide where it’s going to go with Apple’s influence backing it up. But like someone pointed out here… Apple needs competition within the ranks of the PowerPC platform, which I hope that Amiga bieng the newcomer to the platform will bring it on so there is no more excuse that the PowerPC is for embedded use only and that attitude ceases to exist!
Apple is not going to leave PowerPC plain and simple, they wont fall behind the curve ball much more, I think once Mac OS X 10.2 (code name ‘Jaguar’) gets sorted, they can concentrate more on the hardware issues. It all get worked out eventually, it isn’t like Apple has been behind the curve ball that long, but I guess they have to take it one step at a time 🙂
Hmm, the comment about EV6 vs NUMA was a little off base. EV6 is a bus standard, NUMA is a type of processor/memory interconnect. EV6 belongs to a catagory of interrconnects called busses. A bus is shared by each entity on it. With NUMA, systems, however, each processor is connected to its local memory by a bus, and all the processor/memory pairs are connected via an interconnect. Thus, accessing memory from the local bank is fast, but when a processor accesses memory that is in another processor’s local bank, the processor has to do several lookups and go out over the interconnect to get the data. NUMAs have several advantages. When you’re working with hundreds of CPUs, NUMAs scale much easier, since there isn’t so much pressure on a single bus. The disadvantage to NUMA are twofold. First, it’s a pain to program for in a general purpose system. Since not all RAM can be accessed at the same speed, programs have to be aware of where key data structures are allocated. This is especially a pain for those writing kernels, because it makes allocation/management more complex. Second, latency for accessing non-local memory is high. An article at the Register http://www.theregister.co.uk/content/3/22278.html
pegs accesses to non-local RAM at about 140-160 nano-seconds. That seem’s fast, but that’s more than 200 clock cycles for a 1.5GHz processor. Now, compare this to memory access on a regular Athlon machine. On my machine (1.5GHz Athlon XP w/ PC133 SDRAM) my memory allocator (which is mainly limited in speed by how quickly it can make 5 or 6 very spread out (ie. hard to cache) memory accesses) takes about 100ns per iteration. Assuming 100% overhead from memory, that’s about 20ns per access, or 30 clock cycles. Personally, I have no clue why AMD is moving to NUMA for such small machines (4 – 8 processors). My guess is that it probably isn’t NUMA in the classic sense, that AMD has jiggered up some way to make it faster than traditional NUMA for the not-so-many-processors case.
Apple HARDWARE doesn’t make sense anymore.
Worst than that, Apple can´t see the Path they are putting themselves into.
Bad choices in bussiness tend to marginally lose customers.
I never had a Apple but I would love to work with Apple OS X.
Now that AMD is finishing their Hammer/OPTERON 64 bit processor and was using a SuSE Linux OS to run tests, and had to ask Microsoft to produce a 64 bit version of Windows I, a computer user (not an Apple hardware and OS X user), would LOVE to be able to buy an Opteron 64 bit with a wider (DDRam) front side bus AND run a copy of Apple OS X 64 bit !
And, Apple OS X will follow the 4.4 FreeBSD tree so (some, at least) changes will have to be made anyway:-)
I’m looking at Apple’s problem from a very high level. Their problem is not enough people are using it. I’m going to break down the user base into 4 categories: Home users (ma and pa), office users, creative pros, geeks.
Home users: Tasks include web sufing, email, word processing, basically tasks that almost any computer can do. Macs fit in here wonderfully. Wintel boxes are no real threat here.
Office users: Spreadsheets, database, collaboration, more email. Hands down Microsoft Office wins. This disturbs me because I work in an office that’s so dependent of MS Office, and the damn thing breaks more often than it is productive. True, MS Office is available for Mac, but still Wintel boxes are more popular because most offices buy from Dell, Compaq, HP, etc.
Creative pros: Big time Hollywood creative pros use SGI, or maybe Sun and IBM graphics stations. Macs are used in all ranges in the creative industry. But everything you can do on a Mac, you can also do on the PC and vice versa. Recent benchmarks show that Dual Athlons can work harder and cost less than Dual G4s.
Geeks: Up until OSX, there wasn’t much for geeks. We now have Unix wrapped in a MacOS glitter coat. There are more web development tools available as well, and networking is much improved and more flexible. OSX offers more for geeks, but still doesn’t quite hit the spot.
It’s evident that Apple is trying to expand its userbase to not only attract more people, but different types of people as well. We’re starting to see Apple get out of its “niche”. Improvements in the OS is a good start, and the discussion here seems to be about hardware. I’d rather see Apple license PowerPC technology and make it better themselves. Many journalists have praised Apple as being innovators. So lets see them be innovative and roll out their own PPC chips. The saying goes “If you want something done right, do it yourself”. If Motorola and IBM aren’t interested in desktop PPC, Apple should take up the research instead. The rumors we’re hearing about Apple partnering with AMD is more likely to be about fabricating chips, not actually using AMD technology.
To recap, I want to see Apple license PPC and roll out their own chips. If they do partner up with AMD, I’m hoping it’s to get that awesome Alpha team working to improve the PPC design. I don’t know what the Alpha and PPC CPUs have in common, but with an engineering team that good, they can work wonders for PPC. Not only can AMD lend a hand in the R & D, they have their own fab to make the chips. Wishing is all I can do for now.
doubt that Apple and AMD will get together.
AMD is starting to farm out fabing to other companys, so they seem to be maxed out on making chips.
AMD also has licensed MIPS.
Seems AMD thinks MIPS is they way to go for non-x86 processors.
Thanks for doing the comparison. I have been looking to buy a new Apple, and the test was useful. Too bad these kinds of comparisons are like comparing Jesus to Allah in the eyes of the zelaots.
“BTW, it seems that the Mac users feel left behind these days.”
Sad, but true…
I’m sure Apple has something in the works. It just may take awhile for us to get there since the recent new powermac revisions have been in 133MHz steps.
– Mark
Gotta love those mac zealots and how they keep saying they don’t need all that power and don’t want to spend the extra money on it anyway…
Well, it’s not about spending extra money anyway considering a faster pc is cheaper than a g4 system.
So why not save money AND get a faster system? Still don’t
want that extra power? Then why not get an even cheaper pc?
What it’s about is zealousy. For example, Mac zealots liked to say how great their OS was waaay before OS X came out. Of course they ignored the fact that Mac OS < 10 lacked key features such as… true (preemptive) multitasking, yeh yeh, I know the mouse was preemptive but it was the only preemptive part, just like in out of date win 3.1. Even Windows, starting with 95, was preemptive (do NOT misunderstand, I am NOT saying 9x was fully preemptive, but it was partly preemptive with more than just the mouse which still beats MacOS < 10). So Apple didn’t have an OS with true multitasking till they borrowed from another OS (FreeBSD).
It’s not that apple hardware costs so much to produce either, Steve just likes to gouge his faithful customers.
Nothing wrong with that you might say? Well, I suppose it’s not really wrong, but I won’t be taking my money to apple until they get competitive in their prices. Of course, it would also help if the hardware was more open like for the x86 platform. But if the prices were competitive, I seriously consider jumping ship.
So anyway, Mac Zealots, quit complaining about how you don’t need that extra power, the real issue is apple hardware tends to cost much more than an equivilent or even a faster PC. Well, the other issue is the claims made some mac zealots that apple hardware is definately faster for (almost?) everything as opposed to pc hardware.
how much do you think it costs to do the many inovative and artistic designs that apple puts together?
today’s macs are not just POS parts thrown into a crappy case, they are meticulousely designed so as to provide the least amount of noise and the most asthetic apeal.
> they are meticulousely designed so as to provide the least amount of noise and the most asthetic apeal.
I thought Macs were computers, not women.
Too bad it gets so bloddy emotional here @ osnews.
First, this article is kinda off base. Anyone who is paying enough attention to Apple’s marketting machine should realize that what Apple is comparing is thier vector unit to the x86 vector units. I would guess that in the situations where the PPC fell behind the higher clock rate AMD there was little use of the vector units and they are relying on the pure power of the floating point units. So, not a big surprise the higher clock rate x86 beat out the PPC in those cases.
It would also be interesting to see what compilers are being used at Adobe for the two platforms. They are probably using Microsoft’s compiler for the x86 and it generates VERY good code these days. I doubt that codewarrior or gcc can match it on any platform, and I can’t think of any other possible compiler they could be using for the PPC.
Unfortunately ‘zealotry’ goes in both directions from PCdom to Macdom and vice versa. For some reason PC users feel threatened by Mac users and always ask the question why do we prefer our platforms over PC platforms.
“Well, it’s not about spending extra money anyway considering a faster pc is cheaper than a g4 system.”
Lets do a price check, shall we, both systems have similar specs from the top model range…
Dimension 8200 Series
$3,246.00
Power Mac G4
$2,999.00
Both have similar specs from one end to the other, both are the top model in their class (though Apple has one custom build option, but I wanted to compare 2 very closely). That is what is so funny about this PCs are way cheaper with price/performance. Probably slightly cheaper, but not the way you make it out to be. And one benchmark as we saw above is not the only one in town, I just saw Intel advocates (on another website) admitting they can’t beat us on the RC5 benchmark, but that doesn’t really make us king by no means.
As quoted from another forum:
“It used to puzzle me why so many PC users seem to swarm around any discussion concerning Macs and Apple technology. It doesn’t anymore. You don’t see Mac users going around worrying about Compaq’s future, but some PC users just can’t stand the fact that Macs exist and people like using them. I think there is a little niggling fear that maybe, just maybe there’s something out there that’s better than theirs, and that just drives them crazy.”
And you know the really funny part about the whole thing, is that many Mac users use to be PC users (myself included) so obviously this dream world that you try to make up for PCdom. Why does everybody have to be a PC user, are we not allowed to think for ourselves, we’re not normal because we would like to use something that is not Windows (and having some mythical weird jealousy because Microsoft is rich is absurd, try making up a better excuse).
“Even Windows, starting with 95, was preemptive (do NOT misunderstand, I am NOT saying 9x was fully preemptive, but it was partly preemptive with more than just the mouse which still beats MacOS < 10). So Apple didn’t have an OS with true multitasking till they borrowed from another OS (FreeBSD).”
Yeah preemptive multitasking is great until you like to listen to some of your favorite music (CDs, MP3s) and then a skip or 2 that drives you insane, that is where cooperative multitasking seems to shine, so I wont complain in that department. So now Apple borrows from other OSes, why don’t you take a read from this little article for clarity on what runs under Windows XP…
http://www.win2000mag.com/Articles/Index.cfm?ArticleID=4494&pg=1
“It’s not that apple hardware costs so much to produce either, Steve just likes to gouge his faithful customers.
Nothing wrong with that you might say?”
Unfortunately they all like to gouge their customers, that’s what makes the world go round, Microsoft sells MS Office at near gold prices, when there are cheaper alternatives (if not free ones) and the things that make MS Office special most consumers don’t even use! So why do people still pay hefty for that software?
“So anyway, Mac Zealots, quit complaining about how you don’t need that extra power, the real issue is apple hardware tends to cost much more than an equivilent or even a faster PC.”
Well to this day running either Mac OS 9/X my 400 MHz G3 still outperforms our Compaq DeskPros with 400 MHz PIIs and PIIIs running Win2k (which just got upgraded from NT). Though when I bought my first Mac it had nothing to do with the CPU, it had to do with the whole package and I still haven’t found a reason to switch back and probably never will!
PC zealots are just as bad as Mac zealots, and neither of the 2 are going away, so get over it!!!
Geeez, I feel overwhelmed by all your specs. I still think my PII 300 / 160 mb ram is a monster, it even got ATA33 and a TNT2 card!
Everything feels responsive in BeOS / Gentoo Linux and Win2k runs at an adueqate speed too. Sure compilation times are high for things like X and KDE, but otoh blackbox is a matter of minutes – and I do my work faster in rxvt than KDE anyway.
My server is a P100 with 48 mb ram and 2×850 mb IDE disks. Wow – the system load stay ridiciously low running debian stable with IPtables, apache, sshd, proftpd, squid, mail services, QoS etc…
I really think WinXP and MacOSX is the biggest issue when it comes to speed for most of you NOT doing 3D Graphics or Video editing.
And btw – 3D graphics doesn’t look beter now than 5 years ago…
OTOH, of course I wouldn’t by a G4 if I was to buy new hardware today – you always buy what gives you the most bang for your bucks, right? (probably a AthlonXP/MP system these days)
>discussion here seems to be about hardware. I’d rather see Apple
>license PowerPC technology and make it better themselves.
license it? from themselves…they part own it! PowerPC came about because of IBM, Motorola and Apple getting together.
I’ve read from more than one source that the G5 is an Apple design.
Apple also designd Altivec.
>The rumors we’re hearing about Apple partnering with AMD is more likely
>to be about fabricating chips, not actually using AMD technology.
IIRC AMD’s process partner is Motorola so building PPCs in an AMD fab should be relatively straight forward, going elsewhere would prove a lot more difficult.
—
>G5 is good. Problem is, it is too late already
Until we get real performance figures no one can say that, if it’s a completely new generation the performance is likely to be very different to the G4.
Remember the above tests are comparing *systems* in which the CPU is just one component. If apple improve the CPU and the system (which they seem to be doing) this could make a big difference. Ever messed around with a PC’s BIOS settings? These can make quite a difference without changing the CPU settings.
I’m not saying the G4 is better than a PC but low measured performance could be the result of many factors not just the CPU and if Apple can improve things we could get better performance even without changing the CPU.
Besides (as many people montion in these discussions) not eveyone buys a computer on MHz alone, my sister has an iMac (old version), she doesn’t even know what the CPU is never mind it’s speed! Apart from crashing a lot (Mac OS 9…) it does it’s job just fine.
Many buyers simply aren’t interested in the insides of a computer. Just what they do, how they look etc, this market is massively larger than the geek market and this is one area Apple can clean up in.
As for prices – they have such demand for the new iMac they could raise prices without being bothered too much.
—
>I thought Macs were computers, not women.
are you saying women are quiet???
🙂
Hiryu has hit the same nail on the head that I was talking about, and Jeremy is deing exactly what I’m complaining about too. Total user experience and asthetically pleasing designs are definately one reason to choose a Mac over a PC. However, if the equivalent processing power of that Mac costs twice as much, or isn’t available at all, then the asthetics and total user experience becomes irrelevant. Apple is not at that point right now. They are on the edge of the chasm however. They have their OS strategy, software vendor strategy, business strategy and peripheral strategy *finally* up to the task of making some more inroads into the PC user base. If they don’t fix their hardware strategy *now*, then all the other factors become moot, and they are back to where they were in the pre-Rhapsody days.
I would expect the AMD 1.5Gz based PC to blowaway the PPC based 1Gz Apple.
I believe that PCs are better for *me* to use than Macs. There are many reasons for that in addition to Speed, price/performance, OS Choices, Hardware options, ease of use, features etc…
I use to be a MacUser, still have a 7600/132 in a box somewhere. While Apple makes some fine machines, there is no reason for me to use Mac at this time. I used to run BeOS on my 7600/132 but Dano does not run there so…
ciao
yc
doubt that Apple and AMD will get together.
AMD is starting to farm out fabing to other companys, so they seem to be maxed out on making chips.
AMD also has licensed MIPS.
Seems AMD thinks MIPS is they way to go for non-x86 processors.
AMD’s licensed MIPS to make their own embedded processors to compete against Intel’s StrongARM. I agree that it’s doubtful Apple and AMD will work together. I was just hoping for too much.
The benchmarks don’t compare chipsets, they compare applications on operating systems using filesystems on hardware, hardly a basis for making conclusions about hardware.
A more accurate comparison would be to test the same application (e.g., the Gimp) on the same OS/kernel (linux-i86 vs linux-ppc) on different chipsets, though one could still argue that that comparison is unfair, due to better code generation on one platform over another, failure to exploit altivex, and so forth.
Heh…..but Eugenia, for most geeks, their computer is as close as they will ever get to having a significant other 🙂
As I’ve pointed out before, aesthetics count for Apple’s consumer market. They’re explicitly targeting people who are the analogues of those who buy Bang & Olufsen sound systems. The style is part of the price. B&O can’t compete with the stuff you get at Best Buy on price and volume–so they don’t try. Instead, their marketing tactic is to get people to buy their audio equipment at a premium. Moderately low volume is, in fact, desirable for them.
This is something that I think will always be alien to the true geek “market.” If you’re looking for price to performance, move along, nothing to see here. The fallacy is assuming that everyone thinks that way. This persistent fallacy is why Apple’s death has been predicted for almost two decades now. By the same logic, nobody should be buying luxury cars or high-end sports coupes. A Mustang GT is only 0.2 seconds slower than a Mercedes CL and it’s a quarter the price! Face the unthinkable: A strategy which does not emphasize low price and high volume can still succeed.
Having said that, of course Apple needs to get their machines to be faster. I have no doubt that they’re going to, simply because it seems they’re starting to retarget their niche markets–not power geeks, but video professionals who’ve been sending the message, “We’d love to use Macs but we can’t yet.” This is likely to end up giving us speed in the manner of SGI workstations (and, yes, Amigas)–not necessarily faster CPUs, but more processing functions taken over by dedicated processors.
I am surprised that it would come from your fingertips. I hope there are no feminists lurking around, they might take it the wrong way 🙂
My point is not that I think raw performence is totaly irrelivent, but that the total package is better than what you get in a PC, and that is what sells Macs.
the speed of the proc does not hinder me, so I am not totaly concerned with it, that does not mean that I do not worry about the broader implications of falling far behind in the performence of the machines.
the think wit Apple is that they never anounce a product until it is ready to demo. they never say “we are going to develope x system”
that kind of makes people worry becasue other tech companies telegraph their future products.
They talk all sorts of crap about how the mac is better, Ghz doesn’t matter, G4 is a super computer, blah blah. But when obvious benchmarks come out, they immediately say “it’s not about performance!”
I agree with that statement, people buy macs for lots of other reasons. The mac is relatively pretty good in some niche areas especially artistic, it looks good (to some people I guess). Hey, I even almost considered getting an imac on clearance (despite the fact it looked like a plastic piece of junk to my non-artistic eyes) because I have heard and seen such good things about OS X and wanted to play with it.
All those things said – mac maniacs bring this on themselves when they constantly try to pimp apple as being so much better.
I was one of those people who is swayed by the Macintosh user experience, once OS X came out. I even payed too much to have it running on a nice looking neato computer (G4 Cube). However the overwhelming majority of users aren’t in that category. This is a small problem now. It will be a major problem if their architecture doesn’t get up to snuff fast. When I bought my G4 Cube, I didn’t mind paying a few extra hundred dollars for a computer that was at least as powerful as the PC counterparts, faster on floating point operations, and running the neatest operating system I had ever seen.
Fast forward this scenario one year. The G4 is running at 1.2GHz with a 133MHz bus. The equivalent priced PC is running at close to 3GHz with a 500+ MHz bus. To get to the equivalent non-floating point performance, the G4 would have to be running at near 2GHz, and the memory bandwidth would still be lagging behind. If I had to buy my cube under those conditions, I would not have spent that money.
All I’m saying is, that this should be used as a wake up call for Apple and Mac zealots who don’t think there is a performance issue that needs to be addressed in the near term. *Right now* it doesn’t affect sales. If it isn’t addressed, it *will*.
Here’s a list of new G4’s on the way:
http://www.theregister.co.uk/content/archive/24018.html
There’s one due in a few months which should add DDR support a bigger cache and up the clock speed. This and OS X improvements should reduce the speed difference quite a bit.
The most interesting one is the 7500 due next year, it changes the pipeline length to 11( integer) which IIRC is the same as the Athlon and should thus allow for very close clock speeds. As far as I’m aware this is a G4 CPU not the G5, but it could be.
>Face the unthinkable: A strategy which does not emphasize low price
>and high volume can still succeed.
Can and does in many, many different industries. In the Hi-Fi industry there are dozens of small British companies selling expensive but very good kit. Most people wont recognise the names because you don’t see them in normal shops, you just see the normal Japanese stuff there. They are usually minimalist both internally and externally, fewer comonents keeps the audio signal clean, I’ve a pair of British speakers (B&W 602-2) and they’re the best I’ve ever heard, and the only speakers I’ve ever heard which are clearer than headphones. They cost quite a bit, I could have bought an entire Hi-Fi for the same price. I probably could have got a personal CD player for the price of the *cables* alone. Was they worth it? yes, every penny.
I know one guy *a techie* who got a P4 instead of an Athlon, why? Sound. The P4 was slower but quieter. He made a good choice, I got the Athlon …and I’d pay good money for a quiet computer. I would like to hear my speakers over my PC without deafining myself.
BTW. Eugenia, does the Cube have a fan?
—
As for Macs being slow, remember this is Steve Jobs…
Remember the first Next hardware? It was miles ahead of everyone else, blew the Amiga into next week.
But they had exactly the same problem as the current Macs – Display Postscript was slow and people complained. So, they upgraded the hardware with an i860. This was before the Alpha appeared, the i860 – an Intel chip – was the fastest RISC processor in the world (it was also the first 1million transistor CPU and the first to beat a Cray 1 on floating point)
Steve Jobs added it to the Next as a graphics co-processor!
Who knows what Apple have up their sleeves in the next year of two but you can bet it’ll be good! and unlike Next, Apple have volume.
The i860 processor was only available on the high end video card options for the NeXT cubes. The workstations never had such a video upgrade. For information on this video card option see:
http://channelu.com/NeXT/History/NeXTPub/N6030/page4.html
I really, really wish computer users would stop making horribly inaccurate car (and other luxury goods) analogies.
From WattsM:
A strategy which does not emphasize low price and high volume can still succeed.
You’re basing your argument on speaker systems and BMWs. The fundamental problem with this comparison, of course, is that when I buy speakers or a BMW, I don’t care about what everyone else has. It doesn’t affect me; the capabilities of the speakers or the car are based solely upon how they were built. This is certainly not the case with computers, where the capabilities depend completely on the available software. The available software depends completely upon the size of the platform’s userbase! So as a user, it’s very important to me what other people buy.
That’s why low-volume platform sales is usually a death wish: fewer customers = less potential business = fewer companies producing software for the platform = even less software = even fewer capabilities = even fewer customers = … How well do you think those speakers would sell if they didn’t work with audio CDs or DVDs? Or how well would BMWs sell if you couldn’t go to Chevron or wherever, but rather had to fill up only at special “BMW-only” gas stations? How well did Sega Master systems fare? How about 3DOs? Neo-Geos? Sega Saturns? Virtual Boys? When the product has little intrinsic value, arguing that high-cost, low-volume is sustainable is greatly misled thinking.
Rev Aaron:
Maybe I’m a rare breed, a pragmatist. I think of it like a car- I’d much rather buy one that worked just like I wanted it to, didn’t need maintenance very often, got good gas milleage. I certainly wouldn’t base my descision in buying a car on how fast it went, as long as it went fast enough for the roads on which I drive.
This is an even worse analogy. Of course the top speed of a car doesn’t matter to you, because the law limits how fast you can drive it! If you double the speed of your computer, everthing you do is sped up by a factor of two! It’s like if a certain car could get you to work twice as fast somehow, or get you to your best friend’s house twice as fast. Of course, that’s not the case with a new car, which is exactly why your analogy is so poor. Now, maybe that still doesn’t matter much to you. Maybe, you say, it only takes you 30 seconds to get to your friend’s house right now, and saving that 15 seconds wouldn’t be worth the other hassles. And that’s a perfectly acceptable response. But notice how much weaker a statement it is than the original claim you made. For a lot of people, that time will matter. And, I suspect, it will matter to you in the future, just like people who bought “fast enough” G3s two mere years ago struggle to run OS X acceptably.
Just to note it before someone else points it out, I guess I shouldn’t have said “everything you do is sped up by a factor of two.” Obviously, your writing a report, where the bottleneck is you and not the computer, isn’t sped up much. It would be more accurate to say, “everything the computer does is sped up by a factor of two.” It doesn’t change the argument in the slightest, but I thought I’d point it out before someone else does.
“And, I suspect, it will matter to you in the future, just like people who bought “fast enough” G3s two mere years ago struggle to run OS X acceptably.”
Well I guess I must have the one and a million G3 iMacs then, because it doesn’t struggle with Mac OS X. Unfortunately I can’t say the same for Compaq DeskPros running Win2k at work (they just got upgraded from NT a month or so ago)!
when talking about the reasons people buy a car and a computer, you can very well make an analogy. when I buy a computer I look at what I plan to do on it. same thing when Buying a car. I also want one that, to me, looks atractive and is ergonamic. same when I buy a computer.
if it does what I want, then I look at my other criteria.
since most computers do what I want, I rely on look and ergonomics of the software/OS. to me, Apple offers the best solution. and what exactly does an apple do that does not interoperate with the rest of the world?
there is a new processor Archetecture out called XPP. it runs at 100 MHz, but offers more MIPS than P4s.
http://www.theregister.co.uk/content/3/20576.html
What about measuring the power of a Processor/OS by the actual products that people deliver using them? Maybe you will find that PC are fast machines that only a few people can really exploit, and that Macs are in the hands of creative people (creative as in capable of actually producing something worth a dime).
>I really, really wish computer users would stop making >horribly inaccurate car (and other luxury goods) analogies.
This is all very good but you see I’m an ex Amiga user and now a BeOS user. both platforms have bugger all software available for them but I used Amiga for years and use BeOS perfectly happily now so this argument is completely irrelevant to me. There may be less software for the Mac but there is still plenty of it.
>That’s why low-volume platform sales is usually a death
>wish: fewer customers
By that argument there should only be one operating system and thats Windows 98. No Linux, No BSD, No BeOS, no nothing. I don’t think that would be a very good situation and if it was we wouldn’t be having this discussion right now because this site wouldn’t exist.
A luxury good argument is a good analagy for the Mac. Apple have control over the complete experience from the components, logic boards, Operating System, the user interface, the box it’s in and the software it ships with
, they can make it good or they can make it bad. It appears they are making a good system from the comments of most of the people who use it or have recently switched to it.
There is not a single PC company who have even close to that level of control over their systems and consequently you get a system which has been put together from different parts for wildly different companies.
Linux is doing fine.
Amiga is making a comeback,
Apple is doing better than most PC makers…
YAAAAAAAAAAAAAWWWWWWWWWWWWWWWWWNNNNNNNNNNNNNNNNNN!!!!!!!!
MacOS X is great on my G4 but only an Athlon XP can give me the framerates I need playing RTCW. G5 at this time is just vaporware sad to say.
This is all very good but you see I’m an ex Amiga user and now a BeOS user. both platforms have bugger all software available for them but I used Amiga for years and use BeOS perfectly happily now so this argument is completely irrelevant to me.
Yes, there are some people out there who have very limited needs. The vast majority of the population aren’t those people (try explaining to your mother why she can’t use Quicken, or why a website using a particular plugin won’t load, or why the special features on her new audio CD won’t run, or…).
There may be less software for the Mac but there is still plenty of it.
Tell that to any Mac user who wanted to play Half-Life, or wants to use Kazaa, or ICQ (during the years before they finally made a Mac version), or …. ad nauseum. Yes, there is a ton of quality Mac software, enough to fulfill the majority of people’s needs. However, it’s always an uphill fight for them, and the single best way to have more available software is to increase the userbase. Without doing so, you’ll remain behind in available software, and risk losing users (and thus more software) because of it. Thus, companies who think that a small userbase is perfectly fine will soon find that there is no longer “plenty of it.” Thankfully, unlike the apparent beliefs of a lot of Mac users, I don’t think Apple thinks a small userbase is fine; hence, their aggressive campaigns to increase their numbers.
By that argument there should only be one operating system and thats Windows 98.
You’ll notice that I deliberately used the word “platform.” As you might know, Windows 95, 98, Me, NT, 2000, and XP, with very few exceptions, run the same software. Do you think Microsoft could have moved people to the NT kernel so easily had the software been incompatible?
A luxury good argument is a good analagy for the Mac. Apple have control over the complete experience
You say this, but your “justification” for it is completely irrelevant. It doesn’t matter to the argument that Apple has integration capabilities beyond that of any PC vendor. Sure, that means they can offer some very nice things to the user, and I would never dispute that. But guess what? That’s not the point! The point is that with a computer, the size of the userbase matters to each user, and thus any comparison to car companies or audio companies or any other market where that is not the case will inherently be inaccurate. You went off on a tangent rather than answering the point.
Linux is doing fine.
Linux on the desktop has a tiny, tiny fraction of the number of users that Windows does, and its single biggest attraction there is that it’s free (which, of course, doesn’t suggest that it makes for a sustainable business plan, and thus cannot fairly be regarded when discussing Apple’s business plans). Linux and BSD are mainly “doing fine” because they are very good in specialized markets in which much of the necessary software is available for those platforms. Apple targets home users, business users, and media professionals. The first two of those markets, at least, are certainly not specialized.
Amiga is making a comeback
In your dreams. But that’s another matter. For the present discussion, sticking to facts would be preferred, no?
Apple is doing better than most PC makers…
I don’t have the data to either support or combat this (and I suspect that you don’t, either, but that is, of course, speculation). The big question is always long-term viability. I never said that Apple was inherently doomed or any such thing; I merely stated that Macheads claiming that a.) speed doesn’t matter or b.) that a small market is sustainable are gravely mistaken. I used to be a Machead and I know what it’s like to try to deny reality. Thankfully, I think Apple has a better sense of the reality of the situation than most Mac users will allow themselves to see. It will definitely be interesting to watch…
Nicholas Blachford,
Apple did not develop AltiVec. That’s Motorola’s brain child. If you check out the AltiVec developer docs on Apple’s website, they point to Motorola’s website.
I’m pointing this out because other than the cool exterior designs, I don’t know what it is that Apple makes hardware wise.
look in Forbs regarding earnings in the last quarter for computer companies. Dell and Apple are the ONLY computer equipment manufacturers that made money.
look at Apple’s finacial disclosure on and on-line trade site, they are profitable and are sitting on a large quantity of liquid capitol, they have no debt, and move millions of units.
market share does not have anything to do with sustainability.
the mainframe market is tiny, but IBM is making huge profits from it.
Dell and Apple are the ONLY computer equipment manufacturers that made money
And this counters my point how?
market share does not have anything to do with sustainability.
the mainframe market is tiny, but IBM is making huge profits from it.
How on earth can you compare the mainframe market with a general use computer? Mainframes are insanely profitable, there’s little competition, and the machines are tailored to specific tasks. In other words, they don’t require the general appeal that a desktop computer does, and there aren’t many alternatives. You think an airline can say, “Aww, we don’t like this IBM mainframe. Let’s go use a dual Athlon instead!”?
I’d love to hear the executives behind the Amiga, GEOS, OS/2, the BeOS, and countless others react to your “market share does not have anything to do with sustainability” comment. When they stop their laughing, tell them your mainframe “justification.” Just be prepared to treat humor-induced CPR…
what a load of b******s..
yeah, so the G4 is slower.. at least the software for it is better than the totally crap and user-unfriendly windoze shite people have been put up with for so long.
Oh, but usability doesn’t count, right? only numbers.
can we stop doing these complete useless comparisons? no, I do not have a Mac, but I’m fed up with the complete crap the x86 world is spouting.. a system that is still crippled by it’s legacy technology..
point 1)
“Apple is doing better than most PC makers…
I don’t have the data to either support or combat this (and I suspect that you don’t, either, but that is, of course, speculation). The big question is always long-term viability. I never said that Apple was inherently doomed or any such thing; I merely stated that Macheads claiming that a.) speed doesn’t matter or b.) that a small market is sustainable are gravely mistaken. I used to be a Machead and I know what it’s like to try to deny reality. Thankfully, I think Apple has a better sense of the reality of the situation than most Mac users will allow themselves to see. It will definitely be interesting to watch…”
this is what you said, I gave you yout information that you needed.
point 2) there is a critical mass needed for sustainability. the Amiga did not have a userbase the size of the mac, neither did BeOS. OS/2 had a huge userbase compaired to the mac, however, it was in direct competition to MS, and MS pumled OS/2. GeOS floped because just like OS/2 it went up against MS, though it was DR-DOS v. MS-DOS. realy GeOS never got a good hold on the market since DR-DOS was already dieing when Digital released it.
my mainframe justification was more about sustainable market share. as I have said above, there is a critical mass to have a sustainable userbase, and Apple has it. want proof? look at their profit in their quartery reports. a company that does not have a sustainable userbase doe snot operate with profits.
>Creative pros: Big time Hollywood creative pros use SGI, or maybe Sun and IBM graphics stations. Macs are used in all ranges in the creative industry. But everything you can do on a Mac, you can also do on the PC and vice versa. Recent benchmarks show that Dual Athlons can work harder and cost less than Dual G4s.
You better do more reading. Linux is taking over Hollywood fast. Just check these stories out:
http://www.computer.org/computer/homepage/0202/ec/
http://www.linuxjournal.com/article.php?sid=5472
http://newsforge.com/newsforge/02/04/24/1643238.shtml?tid=23
Apple IS trying to get into Hollywood though doing the same things Microsoft is notorious for. They are buying digital special-effects software companies and dropping support for the other platforms and only developing for OS X.
>Yes, there are some people out there who have very limited needs.
>The vast majority of the population aren’t those people
Aren’t they?
How many applications do people use on a regular basis?
I use a web browser, email, word processor and occasionally graphics.
I’ve loads more apps but don’t use them much if at all.
On the PC, Mac, and other platforms people tend to use key applications (Office, Photoshop etc) and these can make all the difference. If a small graphics software producer abandoned Apple I doubt there’d be much effect, if Adobe abandoned them they’d s**t themselves.
>>There may be less software for the Mac but there is still plenty of it.
>Tell that to any Mac user who wanted to play Half-Life, or wants to use
>Kazaa, or ICQ (during the years before they finally
>made a Mac version), or …. ad nauseum.
I think it’s something of an oversimplification to assume market share or sheer numbers of applications dictates a platforms life or lack thereof. In the games console market this is important but price is also important, which is why MS had to cut the price of the X Box.
The PC market has become highly commoditised numbers may be high but prices are cheap so you have to sell a lot to make a profit.
Macs are not the same as PCs and especially since Steve Jobs came back their strategy is to increase this difference. They have put in a new improved OS, GUI, cases and no doubt will improve hardware, sure they may be slower but given they have more demand for the iMac than they can currently supply It’s hardly a problem right now.
You pay more for the Mac but in return you get a better experience, it is in this respect a luxury good. They can only do this because they control every aspect of their system. A PC maker would find it very difficult if not impossible to differentiate itself from another PC maker in the same way Apple does.
Yes, the number of applications matters but it’s not the only thing that matters. Since Macs are more expensive it also means the people using them are more likely to have money to spend and so will be prepared to spend more on the software. This means you can sell less software but still expect to get a reasonable return. This is exactly how the luxury goods market operates. They sell less but still make money.
The analogy may not be *perfect* for BMWs and Speakers but it is good – BMWs have BMW only service stations. Speakers will not work properly with every amplifier – a bad match can sound awful – or worse, if there’s an impedance mismatch you can fry the amp. A good Hi-Fi is like a chain, one bad part and the rest suffers. OK the analogy is not quite application software but there are issues of compatibility as there are with computers.
>>By that argument there should only be one operating system and
>>thats Windows 98.
>You’ll notice that I deliberately used the word “platform.” As you might
>know, Windows 95, 98, Me, NT, 2000, and XP, with
>very few exceptions, run the same software. Do you think Microsoft could
>have moved people to the NT kernel so easily
>had the software been incompatible?
OK then, By that argument there should only be one “platform” and thats Windows. This is not a scenario I warm to.
I have Windows 98 software which will not work on Windows 98 never mind XP, with a complex PC you can run into all sorts of software / hardware mismatch problems. The experience Apple provides means you don’t have to install extra hardware and software and hope it works – in the case of video it’s already provided.
>>Amiga is making a comeback
>In your dreams. But that’s another matter. For the present discussion,
>sticking to facts would be preferred, no?
I guess you’ve not heard about their deal with Nokia or Microsoft then.
Facts indeed…
>>Apple is doing better than most PC makers…
>I don’t have the data to either support or combat this (and I suspect that you don’t, either, but that is, of course, speculation).
Jeremy answered that one.
The big question is always long-term viability. I never said that Apple was inherently doomed or any such
thing;
>I merely stated that Macheads claiming that a.) speed doesn’t matter or b.)
>that a small market is sustainable are gravely mistaken.
Even if the market is a fraction of the size of the PC market whether the Software vendors can make money matters.
Depends what you mean by “small” – small relative to the PC market or small absolute numbers, Apple may have a small relative market share but they still sell millions of computers. Their market share may have decreased but it’s possible their absolute numbers have actually grown.
>It will definitely be interesting to watch…
That I agree with, I think their current strategy is only beginning to pay off, they have issues to contend with (i.e. speed) but these will be sorted out. I think Apple are in a much better position have have a much better strategy than they had, say 5 years ago.
BTW I have only ever owned one Apple product and that was a mono monitor which was used on an Amiga 🙂
————————-
Re:correction
Apple did not develop AltiVec. That’s Motorola’s brain child. If you check out the AltiVec developer docs on Apple’s website, they point to Motorola’s website.
Well, I don’t know if Apple developed it themselves but they were heavily involved.
This information doesn’t come out that much but you see it mentioned now and again in articles.
Apple have been involved in CPUs for some time, The once owned a 33% stake in ARM and the owned (or part owned) Exponential technology – who developed a never used PowerPC CPU a few years back.