Mac vs. PC Digital Photography Comparison Redux

Macmurph writes "Bibble Labs has released a lightning fast version of the RAW image convertor, MacBibble. According to MacBibble creator, Eric Hyman, "MacBibble 3.x is almost 10 times faster than the manufacturers software when converting RAW files under OSX.". Prelimenary tests indicate the Mac may be faster than PCs in RAW image conversion afterall. This calls into question the relevance of the the hotly debated article Rob Galbraith posted just 3 weeks ago and discussed here on Slashdot. Two thumbs up for the PowerPC G4's AltiVec vector processing engine, now being put to work in MacBibble."

Comments

So why exactly do we care about RAW?

[Anonymous Coward]

my intel cpu does xlat faster. gret, huh?

Re:So why exactly do we care about RAW?

[aluminumtulips]

You should be more concerned about your horrific spelling. 'Great' is spelled G-R-E-A-T.

specific code = faster on any system.

[Anonymous Coward]

i'm sure they could speed it up on any system if they wrote a program specifically for that hardware. it would especially be the case if you were using the vector processing units. i doubt the windows versions were using them either since most software isn't compiled designed for the latest and greatest.

Multi-processors

[BlueMonk]

The way I see it, multi-processor systems need to become more commonplace in the PC world. I don't know why they haven't. Is it a cost issue? My assumption is that's how the G4 performs so well, based on the fact that the multi-threading is what gave the program its edge.

Re:Multi-processors

[swb]

I've always thought that exploiting parallelism would be the next step after we hit some kind of practical performance wall in desktop systems. However, I've been saying that for 7 years and the wall just keeps moving, although I kind of think its gotten a little closer.

It's probably a question of economics more than anything else. A 2 CPU system for most end-user applications probably delivers less percentage increase in performance than its percentage increase in cost right now. But up till now its been cheaper to replace a single CPU with a faster single CPU than to invest more upfront in a multi-CPU system -- you have to keep it longer, which means you fall farther behind the current performance curve.

If it became 'standard' to have them, OS and App vendors would be able to deliver a performance jump out of 2 CPUs through better parallelism that would outweigh the increased hardware costs associated.

In the PC world, there's also the historical problem of lack of mainstream OS support for multiple CPUs -- I can't remember if XP consumer even supports it, now that I think about it -- which creates that chicken-and-egg problem. NT4 was a highly marginal 'consumer' OS, Win2k had more reach but still not what the 9x series had and XP adoption has been slower due to people just keeping PCs longer.

I've had a dual CPU system at home for 3 years and I'm not entirely sure I'd replace it with another one once I looked at the economics of it. The biggest single benefit I can think of is that it doesn't bottleneck the way a single CPU can when a single process pigs out at 100%, I still have a nearly-idle CPU to work with -- which is the problem with 2 CPUs, one's nearly idle.

Re:Multi-processors

[kinnell]

The reason the G4 performs so well is the altivec engine - it has a 128 bit data path which can process 4 32 bit value simultaneously, or 8 16 bit values and so on, as well as having a bunch of useful vector operations. This all gives it a huge edge over the x86 architecture for signal processing applications, such as imaging.

Re:Multi-processors

[Proc6]

As a user of exclusively dual CPU PC's since the P2-300, I have decided I will switch to HyperThreaded single CPUs now that the 3ghz HT is out. People dont understand the benefit of dual untill they use one for awhile. Even a dual 500Mhz PC is far smoother and more productive in general than a single 2.4Ghz proc, unless all you do is play games. On a single proc PC, using one application that requires some CPU attention just brings the whole machine to it's knees. If you haven't seen it yet, check out Tom's Hardware video showing HT vs non-HT head-to-head. It's really enlightened alot of friends and family as to the value of 2 CPUs or HT vs 1. Well worth the extra cost, though I am patiently waiting for the 3Ghz to drop from its excessive $650 for just the CPU.

Re:Multi-processors

[hcdejong]

Could it be that there's also an element of laziness on the programmers' part? I expect it to be easier to write an application that expects to run on one processor (you don't need to worry about dividing tasks over multiple processors to optimize performance) than a multi-processing app.

And, who really tries to optimize performance today? IMO many programmers expect Moore's Law to take care of the performance increase (relative to the previous release of their program). I rarely see a version n+1 of an application that's faster than version n was on the same hardware.

Re:Multi-processors

[cide1]

Ive had a dual cpu for 4 years, and I definitly would not replace it with a single cpu. My dual 700's feel much faster than my 2.2 GHz laptop (actually has a desktop chip in). Any type of file i/o or running more than about 4 apps just slows down the single processor too much. I love being able to run vmware and giving it a whole processor, or when I compile large projects, I can compile two files at once. Grip takes advantage of a second cpu when ripping music. The normal linux system alone, without adding a bunch of user apps has enough threads to see the differance. I agree with earlier posts that there will at one point be a practical wall of performance, Im not sure how far off that is, but parallel processors is great method to work around it.

Re:Multi-processors

[vidarh]

Of course that is based on the fact that most apps are single threaded apps that won't manage to hog more than one CPU at a time. If SMP or HT becomes the common way of increasing speed at some point, then more applications will be heavily threaded and able to exploit this.

Essentially, the effect you're mentioning could be handled on a single CPU machine simply by running a scheduler that guarantee that no process will get any more than every second timeslice, or similar, penalizing single threded applications.

Re:Multi-processors

[Anonymous Coward]

XP Pro has multi-processor support. Most people who are buying a multi-processor computer would never use XP home anyway.

Re:Multi-processors

[sparkhead]

Even a dual 500Mhz PC is far smoother and more productive in general than a single 2.4Ghz proc, unless all you do is play games.

Nice usage of subjective terminology, but while I could believe a dual CPU system outperforms a single CPU system that has twice the clock speed, I find it difficult to believe it would outperform one five times its speed. Post some numbers.

Re:Multi-processors

[swb]

The reason I'm somewhat skeptical of the value (beyond the its-still-a-smooth-UI-when-a-single-task-is-at-100 % reason) of this is my direct experience at home/work.

At work I have (because I'm too cheap) a 1.xx GHz PIII and at home I have a dual PIII 667 (actually 650 overclocked a tiny amount). I think the work PC is faster overall and just as responsive, even though its only a single CPU system.

I realize that much of the advantage is 133 vs 103Mhz bus, but that's part of the point -- if you stick with cheaper, simpler single CPU systems you can get a new one faster and take advantage of other improvements on the motherboard faster than you can with a more expensive (and hence kept longer) dual CPU system.

Re:Multi-processors

[Eccles]

Could it be that there's also an element of laziness on the programmers' part?

It's not laziness, it's priorities. Optimization is low priority in programming; if there's other things that need to be done, they need to be done first. And hardware optimization comes even lower on the priority meter, especially hardware that only a few users have, and especially hardware that will at most give you a 2x speed-up.

Re:Multi-processors

[hcdejong]

Good point. The question is then, why is speed such a low priority when the #1 argument to buy a new PC is that 'it's faster', and the #1 comparison statistic (useless or not) is the processor speed?

Re:Multi-processors

[Anonymous Coward]

wath his impying isnt that the 2x500 computer is faster than a 2.4 ghz computer its just smother.
forexampel lets say you are converting mpgs to divx or doing somthing similar. thats often just a singelthreaded app and it will consume almost all the power from the singel processor machine. so it would be har to do something usefull with the machine in the meantime.
the you try to do the samething on the smp system
sure the task converting the mpg 2 divx vill take 5 times longer but you will still have a free cpu so you can use the computer and for exampel play unrealtournament in the mean time.
the first option leave the computer alone for one houer and grow fustrated.
sec option ok so it takes five houers insted of one but who cares u can still use the computer as normal

Re:Multi-processors

[Anonvmous Coward]

"In the PC world, there's also the historical problem of lack of mainstream OS support for multiple CPUs -- I can't remember if XP consumer even supports it, now that I think about it -- which creates that chicken-and-egg problem. NT4 was a highly marginal 'consumer' OS, Win2k had more reach but still not what the 9x series had and XP adoption has been slower due to people just keeping PCs longer."

XP Home does not support dual processors. As a matter of fact, this may render it an unusuable OS for HT processors because what they do is trick the hardware into thinking it has two processors.

There's something else to consider about OS support for multiple processors: The UI needs to support it, too. I run Win2k on a dual machine, and because I'm aware of how to direct a process to an idle CPU, I'm able to make reasonably effective use of the second processor. Unfortunately, though, I have to go into Task Manager and hunt down the process I want to change. I'm able to use it, but I have a difficult time explaining this to other people. The big problem is that it's hard to tell, at a glance, which process is doing what.

I really hope that one day Windows (or whatever OS I end up using in the future) puts something in the titlebar of each app I'm running so I can set which processors it is active on. That would seriously be cool. I'm not a big fan of running apps in multi-threaded mode because I like having the CPU resources to keep the interface etc going. If, at a click, I could say "stop using this processor", I'd be one happy SMP adopter. I'd even be able to recommend this type of machine to less tech saavy people.

I gotta ask, though, does anybody know of a hack for Windows that let's me do this?

"The biggest single benefit I can think of is that it doesn't bottleneck the way a single CPU can when a single process pigs out at 100%, I still have a nearly-idle CPU to work with -- which is the problem with 2 CPUs, one's nearly idle."

True dat! Explorer is well multi-threaded, and is always very responsive on a dual machine. When I first started working on a dual I couldn't believe how enthusiastic Explorer and IE were, even when running a rendering in the background or something. The dual processor support didn't buy me 2x the speed (though potentially it could get close, depending on how efficient I plan on being...) what it did buy me was that my computer was ALWAYS useful, as opposed to having it completely busy chewing away at something.

Ah how I ache for a dual processor laptop.

Re:Multi-processors

[Anonymous Coward]

Because speed of software is only a priority when your software is competing against similar software. Very few people buy Photoshop because its faster than CorelDraw, but a lot of people buy a 3Ghz PC because it's assumedly faster than a 2Ghz PC.

So long as software speed is perceived as "acceptable", that's generally where the optimization stops. This, of course, is not true for highly competitive software markets. Maybe we need a little more competition.

Re:Multi-processors

[Atzanteol]

Any type of file i/o or running more than about 4 apps just slows down the single processor too much.

FYI: Laptops tend to have much slower HDD's than desktops. This may cause more of a performance hit in this area...

Re:Multi-processors

[1g$man]

XP Home does not support dual processors. As a matter of fact, this may render it an unusuable OS for HT processors because what they do is trick the hardware into thinking it has two processors

Wrong.

XP does not support dual physical processores. However, it fully supports a hyperthreading CPU and it's second logical CPU.

Re:Multi-processors

[Anonvmous Coward]

"XP does not support dual physical processores. However, it fully supports a hyperthreading CPU and it's second logical CPU."

He's right. I found some info on it here.

What it basically says is that XP Home and Pro will both support this processor, but 2K won't.

My company has a couple of HT machines here and when the POST comes up it says '2 processors'. Since we Linux on that machine, we never looked up Windows compatibility.

Re:Multi-processors

[Anonymous Coward]

I've been told that (at least on OS X) it's very easy to make use of SMP from a programmers viewpoint - it's more or a less a couple of include files and a compiler switch.

Re:Multi-processors

[damiam]

Of course it wouldn't win at Divx encoding, but for general desktop use, it quite possibly would. Imagine this: a CPU intensive task is running. Maybe a program's starting up, maybe you're compiling something, or maybe something is locked up and using 100% CPU. That will create moments when the computer is unusable - even on a Thz machine. On a dual machine, there are no such moments, so the computer is always usable and productive.

Re:Multi-processors

[Anonymous Coward]

"What it basically says is that XP Home and Pro will both support this processor, but 2K won't."

Only valid for W2K (workstation).

W2k pro (server/etc) can use both cores (HT/dual).

Next MS limit is 4 cores (2xHT/quad) ;-)

Greetings from Germany.

-Dieter

Re:Multi-processors

[drsmithy]

The way I see it, multi-processor systems need to become more commonplace in the PC world. I don't know why they haven't. Is it a cost issue?

No, it's simply unnecessary - SMP PCs are easily and readily available, they are just "uncommon" because there's no need. Processors on the PC side have increased in speed consistently and reliabily since the days of the Pentium 100 (ca. 1996). Every year brings a much faster CPU at a reduced cost. This is not true on the Mac side - the G4 has run into a brick wall and simply isn't getting much faster (and is doing that slowly). Thus, Apple have reached for the crutch of SMP to keep their machines at least playing the same game (albeit not really in the same ballpark). SMP is certainly there on the PC side if you need it, but by and large you don't - for most people a single, blazingly fast CPU is just more beneficial than a couple of much slower CPUs. Certainly, if you are heavily multitasking and/or running heavily threaded applications then an SMP system will feel more responsive, but for most this simply doesn't apply.

There is no real reason for multiprocessor machines to become "more commonplace". For most people they offer little benefits and for those who do benefit, they just go out an buy one.

My assumption is that's how the G4 performs so well, based on the fact that the multi-threading is what gave the program its edge.

G4s, at least in the manner they appear in PowerMacs, have an atrocious SMP implementation. They are bus-starved and really aren't running anywhere near as fast as they could. The G4s "perform so well" in areas that they do, because of applications specifically tuned to the benefits of the CPU - ie "AltiVec". In general usage the G4 Macs are (relatively) slow. They still haven't gotten to the "fast enough" stage such that they can easily overcome the overheads of the OS (a state PCs reached with Win2k around the 800MHz P3 mark and with XP about the 1.3GHz mark) and as such feel even slower than they really are (deliberately inserted delays in the OS don't help much either).

In short, G4s are pretty fast for their clockspeed, no argument, but their clockspeed is simply *so far behind* than putting them into SMP configurations is the only way to keep them even marginally competitive. Trying to say a 1.4Ghz G4 is "faster" simply because it can keep up with a ~2.2Ghz P4 is meangingless when the G4 can't go much faster at all, and P4s are likely (and designed) to clock to much higher clock speeds than they are already at. If intel had only just released a 2.5Ghz P4, and were having trouble getting them faster then yes, the G4 would be a definite competitor. But they haven't and aren't, so it isn't.

Re:Multi-processors

[Proc6]

I know. My main workstation was an SGI INDY for 2-3 years. IRIX is like that, and I loved it. I could start up 3-4 big render jobs and they'd run in the background, yet you'd not even feel it in the interactive desktop. I loved that. The only real let-down going to WindowsNT (for me) at the time, was that smoothness. Which is why I ended up going dual CPUs.

Re:Multi-processors

[Anonymous Coward]

Web browsers, for example, are far slower on the Macintosh platorm

Quite right. Most of the people buying multiple-processor computers are using Mac OS X.

Re:Multi-processors

[MikeUnicode]

It's bad tools too. MS VB is very bad at multitasking. Most VB programmers arn't bright enough to handle the concept. But, the few that can are hampered by the language itself. Win2000 can't handle lots of VB threads either. It simply lets some fall asleep and never awake.

Re:Multi-processors

[Eccles]

Good point. The question is then, why is speed such a low priority when the #1 argument to buy a new PC is that 'it's faster'

A faster processor optimizes everything, and (assuming Intel/AMD/Motorola/whoever has done their job right) does so without compromising stability. Intel et al can optimize for performance because they've otherwise solved the x86 problem, it works and it works well.

It's not that applications should never be optimized, it's just that it should be lower priority than other things, and an optimization that few can use without a particularly high upside should be especially low priority.

Incredible!

[gazbo]

Man writes fast code on Mac for doing a single task; wannabe journalist extrapolates this to figure that Mac's are faster than PCs.

Stay tuned for next week when we take the existence of bikini waxes to show that George W Bush is a paedophile.

Re:Incredible!

[papasui]

Prelimenary tests indicate the Mac may be faster than PCs in RAW image conversion afterall. I know your joking, but I feel a /. beginning. From the above quote it looks like it's pretty specific on what the Mac is said to be faster at, it doesn't say that it's faster at everything.

Re:Incredible!

[JanneM]

Of course, there is no saying what speed increase a PC would get with a similarily optimized dedicated app for the same task. This proves nothing either way.

Re:Incredible!

[Anonymous Coward]

Macs are faster in most algorithms with source available.

Typically the PowerPC (seen in most of the the www.top500.org list of fastest clusters) trounces Intel and even AMD at almost every benchmark.

Not just the 10 famous benchmarks as part of the composite in ByteMark , but at many other things such as the RC5 contest.

according to the RC5 benchmarks AMD is far slower than dual cpu macintoshes (half as fast). (source available for cor rc5 loops for most
processors)

The Mac Dual 1 Ghz g4 is faster than all existing dual AMD motherboards in RC5 benchmark by almost 100%.

21,129,654 RC5 keyrate for dual 1 Ghz g4 system ! And Now apple sells dual 1.25 Ghz stock and this week a 1.45 Ghz which would be even faster.

A dual 1800+ AMD MP get only HALF as many as a Mac! 10,807,034 rc5 keys !

Funny "Mhz myth" there showing itself I guess... Apple now is selling even FASTER machines than that one I mentioned made over one year ago, but with smaller caches and less fast read-write ram (it
now uses DDR on newest boxes).

The mac I mentioned uses a 2 MB L3 cache and no amd mp dual cpu boards I know about have any L3 cache at all, so maybe that is why some common macs are
over twice as fast, its not just altivec meager tweaks to rc5. AMD have similar , but less mazing vector ops.

The Pentium 4 takes many cycles (over 7?) to do a simple left shift. That is why the Pentium is MUCH slower than even the AMD or Mac.

Most modern CPUs can do a left integer shift in 1 cycle, any barrel position, not 7 slow cycles.

(Shifting is used a lot in decryption, encryption, graphics processing and many things).

Another reason the mac might be over twice as fast as an amd dual mp board is not just the 2MB l3 cache but the fact that mac can read and write to
a cold page of memory simulatneously FASTER than any AMD MP designs which are biased for linear access and streaming. Many memory scatter
benchmarks show this too. Apples newest DDR-RAM machines might not offer this feature though.

True, RC5 fits in primary cache of most machines, though interrupt services need larger caches depending on interrupt designs and load for the rest of the OS.

The RC5 benchmarks are never run with interrupts off, they use real world overhead.

The Macs made since september also can RAPIDLY service every pci slot almost simultaneously one 32 byte cacheline each if needed. How can it do that ? Three cool features of modern PCI :

* out-of-order completion
* address bus streaming
* intervention

Out-of-order completion allows the memory controller to optimize the data bus efficiency by transferring whichever data is ready, rather than having to pass data across the bus in the order the transactions were posted on the bus. This means that a fast DDR SDRAM read can pass a slow PCI read, potentially enabling the processor to do more before it has to wait on the PCI data.

Address-bus streaming allows a single master on the bus to issue multiple address transactions back-to-back. This means that a single master can post addresses at the rate of one every two clocks, rather than one every three clocks, as it is in the 60x bus protocol.

Intervention is a cache-coherency optimization that improves performance for dual-processor systems. If one processor modifies some data, that data first gets stored only in that processor's cache. If the other processor then wants that data, it needs to get the new modified values. In previous systems, the first processor must write the modified data to memory and then the second processor can read the correct values from memory. With intervention, the first processor sends the data directly to the second processor, reducing latency by a factor of ten or more.

ALtivec is not usually the reason a mac performs better than Intel in benchmarks of properly compiled code, because the famous set of 10 algorithms in ByteMark were not using ANY altivec instructions.

And the AMD bests the Intel at Rc5 mainly from integer features.

I laugh when pc people try to dismiss the fastest machine (Macs) by claiming Altivec "cheating" all the time. The mac people should be the ones to call foul when Intel was cuaght PAYING adobe to slow down filters in one version of Photoshop to artificially make the Pentium MMX 166 Mhz look faster. They got caught paying big bucks. Adobe replied that it was an unfortunate side effect of adding optimization for MMX and not keeping the code efficient in the non MMX case as it was before. HA!

Almost every pc person likes to use benchmarks that use lots of assembly for intel (Quake, etc), but shy away from benchmarks that offer source code in ANSI C.

I knew the mac handled RAW better than PCs and this news is no surprise to me.

Re:Incredible!

[Anonymous Coward]

Typical AMD fanboy who hates the truth huh? Steve Jobs promotes amazing technologies in SOFTWARE such as the NeXT Step OS, Mac OS X, and the famous January 1984 Macintosh.

Steve Jobs also promote technologies 3 to 5 years ahead of any pc company. Now makes use 54 megabit wireless stock! 802.11g and 800 MBit Firewire 1eee 1394a.

But you , amd fool, do not understand that coincidentally, just by good fortune , RISC chips are used in workstations and Macintoshes (since 1995) and RISC beats the pants of CISC in certain tasks, such as running easily compiled ANSI C code.

RC5 is TWICE as fast on a Mac than the fastest AMD dual CPU... enough said.

Re:Incredible!

[Nugget]

I don't mean to rain on your enthusiasm and I sure don't intend to imply that I dislike Macs. I'm typing this up on my PowerBook.

Using RC5 as a benchmark is only relevant insofar as you want to compare RC5 processing speeds. There RC5 algorithm, as well as the specific implementation found in dnetc, contain many aspects which make the results you obtain insightful for general use. You simply cannot compare RC5 rates and hope to extrapolate or project them into general processor comparisons.

The RC5 algorithm relies heavily on bitwise rotates (left, if you're curious, ROTL) which is an operation that is not commonly found anywhere outside the world of RC5. This instruction is so underused, in fact, that many x86 architectures (AMD's K6 for instance) have taken to simply emulating the ROTL operation and eliminating true hardware support. This is why some conventionally powerful platforms (such as Sparc and Alpha based systems) do abysmally in RC5 as compared to x86 platforms containing a hardware ROTL implementation.

Then again, this level of detail is probably lost on someone trying to compare a 1GHz G4 against an "AMD motherboard". AMD has made quite a number of CPUs in the past few years and their range of performance on RC5 is very broad. At one time, the AMD K5 was, in fact, the best-performing architecture in RC5 with the most keys per clock. AMD doesn't make any motherboards as far as I know.

The core of dnetc is also small and lean, often fitting entirely in L2 cache on many architectures. This means that dnetc does not adequately (if at all) exercise memory bus bandwidth. The cores also tend to be hand-tuned assembly, so they aren't as likely to exercise a processor's speculative execution routines. RC5 uses absolutely zero floating point math, also an uncommon scenario and not representative of many apps you would traditionally think of when you think of apps which require strong CPUs to perform well.

Many people enjoy having machines which perform well at RC5 and generate impressive distributed.net stats. Consequently, RC5 shows up as a metric in a great number of reviews and analyses on architectures and CPUs. I'm tickled whenever I see it and I think it's a great addition to any CPU review. However, it's not valid to try to make the claim that RC5 performance rates mean anything more than RC5 performance rates.

Moo!

Re:Incredible!

[BWJones]

Actually, this shows the results of system specific optimizations. For instance, molecular modeling code optimized for my old SGI Octane was rippin fast. Much faster than on any other platform I could find. However, code not optimized for the SGI platform was just as fast on Intel or PowerPC. Now, that said, the G4 does have something called Altivec, and code optimized for this can be unbelieveably fast. Optimized BLAST libraries are faster on my dual G4 than anything I have ever used including some big SGI iron.

The trick is getting programmers to take the time and effort to optimize for specific platforms. This takes time and money to write quality code, but in the era of Microsoft timeline driven products, quality software code is harder to come by.

Re:Incredible!

[lederhosen]

How many of of the computers in www.top500.org
is macs? *zero* They are IBM:s with Power3/4
with huge caches, often dual cores. They do
*not* exist in Macs and are so fast mainly because
they exist in fast interconnect networks, i.e.
supercomputers. Macs powerPC:s are made by motorola and are *slow*.

Re:Incredible!

[acidblood]

I won't comment on your entire post, but given that you were quite misinformed about the whole RC5 issue, I bet you are just a very uninformed karma whore.

Let's start...

Macs are faster in most algorithms with source available.

Uhh... I cannot even start to debunk this. Probably because I don't get what you mean, except that you were whoring for karma with the open-source crowd.

Not just the 10 famous benchmarks as part of the composite in ByteMark , but at many other things such as the RC5 contest.

Never heard of them. How about the industry-standard SPEC benchmarks? Oh, wait, Macs are twice as slow when compared to Pentium IIIs with the same clock speed, IIRC. Apple is so ashamed of the processors they use, you won't see a single SPEC benchmark published by Apple.

according to the RC5 benchmarks AMD is far slower than dual cpu macintoshes (half as fast).

I have covered that extensively on the Slashnet forum with DCTI. To make a long story short, the rotate operations (not bit shifts) were made available on the Altivec instruction set, and MMX/SSE2 doesn't have them. Observe that these useless (for the most part) instructions are only provide on the x86 and PowerPC ISAs, all other major CPU architectures do not offer these instructions. The more I think about it, the more it seems Apple was going for ultimate RC5 performance by including these rotate operations on Altivec -- so they could have at least one benchmark they'd always be ahead of everyone else, as long as they can keep their clock speed within 33-50% of x86 processors (that's 2-3 times less, if you haven't realized).

The Pentium 4 takes many cycles (over 7?) to do a simple left shift.

Wrong, only 4 cycles.

Another reason the mac might be over twice as fast as an amd dual mp board is not just the 2MB l3 cache but the fact that mac can read and write to a cold page of memory simulatneously FASTER than any AMD MP designs which are biased for linear access and streaming. Many memory scatter
benchmarks show this too. Apples newest DDR-RAM machines might not offer this feature though.

This has to be the worse piece of BS I have ever read on my life.

Intervention is a cache-coherency optimization that improves performance for dual-processor systems. If one processor modifies some data, that data first gets stored only in that processor's cache. If the other processor then wants that data, it needs to get the new modified values. In previous systems, the first processor must write the modified data to memory and then the second processor can read the correct values from memory. With intervention, the first processor sends the data directly to the second processor, reducing latency by a factor of ten or more.

This is where you have shown how you don't understand anything you're talking about. This cache-snooping protocol is a feature of the Athlon (I doubt the Macs have it), and it is valid for the whole range of memory and not only the PCI bus -- which probably is marked as uncacheable in the MTRR so reads and writes are not cached, you obviously don't want that for I/O data.

Quit the karma whoring, troll.

Re:Incredible!

[Anonymous Coward]

Quite a few of them are not just Power3/4 many are PowerPC in the list. You are a liar. Plus the New chip apple is using in August 2003 **IS** the Power4 in one and 2 core mode and chip sampling starts in March.

Also if you divide the number of CPUS into the top500.org indexes you will see the Power and PowerPC is very efficient per processor! (check 2001 lists when it dominated)

Simple fact is that a one year old mac is twice as fast as the fastest AMD at RC5.

I admit that this year is the worst year for PowerPC in the top500.org , but what do you have to say for earlier years... or NEXT YEAR when the new Power4 "light" ships and Apple uses it in STOCK MACHINES!!!

Re:Incredible!

[Alan Partridge]

it DOES prove one thing - that Macs ARE competitive, and that there's more than just Wintel's game in town. This is great news for everyone even remotely interested in computers.

Re:Incredible!

[JanneM]

Um, no, it doesn't. If the PC-based utilities were written as ineffectively as the mac code was (not an unreasonable assumption), then we're back to the same conclusion as the original article made.

Just to be clear, I'n not slamming Mac or anything here; I like them. I'm just saying that this really does not prove anything at all until a similar, optimized piece of software has been tested on a PC. And even then it proves very little beyond each machines ability to perform in this very specific task.

There's an old truth that the only benchmark that counts is the performance of your applications. This stands today as it did before. The new thing today is that for a lot of garden-variety computing added performance really doesn't matter anymore.

I very recently got myself a laptop which I'm running Linux on. It needs ACPI to handle CPU throttling, which I have not yet patched in. Just a couple of days ago I looked through the startup log and discovered that the bios sets the cpu in full-speed or power-saving mode at startup depending on whether it's connected to a power supply at startup. This setting is then kept for as long as the machine is running. It's a clock speed difference of 1.2 Ghz vs. 1.8Ghz. Thing is, I never noticed this. The machine has felt just as fast and as respponsive whether I happened to start it on battery power or with the mains adapter. Had I not looked at the logs I would never have suspected it.

The truth is, what's limiting computers today have more to do with disk speed and memory management strategies than with cpu speed. Yes, in some cases CPU speed does matter, but even with such applications, most of the time is generally spent waiting for other stuff once you lift the perspective from just one component of the task at hand.

Re:Incredible!

[Anonymous Coward]

Uhm, isn't Bibble avaialable on mac AND PC?
If so, isn't it safe to assume that the code is optimized for both platforms?

And further, if this is true... Then wouldnt this be afair test of both platforms?

Re:Incredible!

[notfancy]

Man writes fast code on Mac for doing a single task; wannabe journalist extrapolates this to figure that Mac's are faster than PCs.

Don't tell me that the quest for the meaningful, objective benchmark has ended in success.

Re:Incredible!

[pi radians]

Never heard of them. How about the industry-standard SPEC benchmarks? Oh, wait, Macs are twice as slow when compared to Pentium IIIs with the same clock speed, IIRC. Apple is so ashamed of the processors they use, you won't see a single SPEC benchmark published by Apple.

I went the specmark.org to look for the proof of your statement, but they actually don't have any motorola processors listed as tested. I would love to find the comparisions that you mentioned. I do remember one instance where Apple actually released SPEC benchmarks where the G4 beat either a P3 or P4 (I can't remember exactly) but there is zero chance on my finding that info now.

Quit the karma whoring, troll.

You tell'em buddy.

Re:Incredible!

[Luminous Coward]

The Pentium 4 takes many cycles (over 7?) to do a simple left shift.

On the Intel NetBurst micro-architecture, SAL, SAR, SHL and SHR have a latency of 4 cycles and a throughput of 1 i.e. you could execute 4 SHL in 7 cycles if they had no data dependency (pipelined execution).

Re:Incredible!

[Anonymous Coward]

That was a pretty funny troll.

Re:Incredible!

[Anonymous Coward]

No.

Re:Incredible!

[Anonymous Coward]

How about the industry-standard SPEC benchmarks? Oh, wait, Macs are twice as slow when compared to Pentium IIIs with the same clock speed, IIRC. Apple is so ashamed of the processors they use, you won't see a single SPEC benchmark published by Apple.

IBM has published some performance guesstimates of the upcoming 1.8 GHz PowerPC 970. SPECint2000 = 937 and SPECfp2000 = 1051. I have no idea how they built the system around the chip.

Re:Incredible!

[Anonymous Coward]

Jobs' software's ok. NeXT was adequate, but besides his graphics rendering, was WAY OVERPRICED and NOT WORTH IT. OS X is getting better, but it's not enough to keep Apple afloat.

Look, shill, IBM could make 10 million ghz processors and 30234 megabit wireless a standard on their PC's, IF THEY THOUGHT ANYONE WOULD BE STUPID ENOUGH TO PAY THE *HUGE* PREMIUM ON IT!!! Yes, the technology on Macs is 3 to 5 years out there, *for a reason*!

Only Mac users are gullible enough to APPLAUD a company making expensive Gigabit ethernet stock when so very few have networks capable of using its full potential. And what's up with this 802.11g networking? Do you know how fast your cable modem is? How much of that 802.11g's bandwidth is left over? The infrastructure's not there to support it. You'll upgrade before you get to really use any of this stuff. And I love the widescreens on the new PB's, but they're not actual widescreen aspect ratio. Widescreens are 1.78, this is 1.6. WTF? And you wouldn't need the extra room for your apps if damn Apple would put back in the multiple terminals every other Unixy product in the world has. Damn Crippled Unix. You're getting shafted, paying for *way* more accessories than you need, but with a low powered CPU that will have you upgrading in a year, before you could even *find* a network that uses Gigabit ethernet.

RISC is *theoretically* a good design. But Moto's real-world implementation is lacking. And for god's sakes, man, the speed advantage is about *a maximum* of 30% overall over CISC. That means a 3Ghz CISC still kills any G4 out there. They'll probably kill the G5's, too.

Apple is a speck on the PC world's radar. A SMALL speck. And unfortunately, they probably WILL go under soon. Even with the coolness of OS X and their new hardware, they're losing marketshare. They're getting squeezed to death because they don't run Windows. They're not compatible with Windows. They don't run Windows software. They have limited hardware choices. They are after a niche market that is being attacked from all sides. We switch because we hate MS, but I hear more negative stuff Apple's doing than Microsoft. Apple is getting just as bad. And Windows XP and 2000 just don't suck that bad anymore, and in a couple years, Linux is going to blow all of the competition away, what with every country and tech company in the known universe developing for it.

RC5 is twice as fast. Whatever. The other 3000 tests are a CISC landslide. Anyway, my "Hello, World" beats your "Hello, world". And it runs on 95% of the world's computers.

Re:Incredible!

[Anonymous Coward]

Agreed 100%. However, time and time again, we've seen that no matter how many statistics you show, no matter how much outright *proof* you have that A is better/worse than B, in the end, one thing is always true: mediocrity wins.

(w)Intel wins over Mac, SPARC, Alpha, whatever. Linux wins over BSD, Solaris, AIX, HPUX, whatever. Windows wins over Linux and Mac. I could go on for days with examples of how a third-rate, consumer-grade product will always whack the superior competition.

You get what you pay for (sometimes).

Re:Incredible!

[Anonymous Coward]

Too bad RC5 is the _only_ benchmark Macs ever win. They're even slower in photoshop, now. Anyway, if you wanted to run a real benchmark, run a SPEC or something, and watch the Mac get its ass handed to it.

Nice fanboyism, btw.

Re:Incredible!

[Anonymous Coward]

What you are neglecting to mention is that most Macheads claim, and rightly so, that the price of similarly equipped Apples and PCs are nearly identical. Since we know that Apple charges a higher profit margin on their goods, and they're priced similarly, Apple must be using cheaper quality components in their hardware to account for the difference. Which would explain why they use last year's memory subsystem, slow bus rates and hard drives, cheaper quality LCD displays, etc. They even skimped on the Titanium PowerBooks' paint, if you'll remember the rash of flaking, chipping PowerBooks....

Bitterman

Re:Incredible!

[PCBman!]

you can find them on JC's cross comparison, they're old though
JC's Home Page
Scroll down, on the right, under Benchmarking.

Re:Incredible!

[Anonymous Coward]

Uhh, Titanium is bad at holding ANY kind of paint. I don't think that the cost of the paint has nothing to do with it.

The Apple LCDs are some of the finest on the market.

You are a TROLL!

Re:Incredible!

[Twirlip of the Mists]

And I love the widescreens on the new PB's, but they're not actual widescreen aspect ratio. Widescreens are 1.78, this is 1.6. WTF?

"Widescreen" doesn't mean anything at all. My computer screen is 1.6:1. My TV screen is 1.78:1, though most of the content I watch on it is 1.85:1. When I go to the movies, the screen is usually 2.35:1. "Widescreen" can be applied with equal truthfulness to any or all of these.

And you wouldn't need the extra room for your apps if damn Apple would put back in the multiple terminals every other Unixy product in the world has. Damn Crippled Unix.

The only thing I can figure is that you're talking about virtual desktops. Either that, or you're an idiot. Maybe both.

You're getting shafted, paying for *way* more accessories than you need, but with a low powered CPU that will have you upgrading in a year, before you could even *find* a network that uses Gigabit ethernet.

My home network is 100% gigabit Ethernet. All you need to build a 100% gigabit Ethernet network is two Macs and a cable. It doesn't even have to be a crossover cable; all Macs are equipped with autosensing MDI-X ports.

That means a 3Ghz CISC still kills any G4 out there.

Except running Bibble, evidently. And BLAST. And all the other stuff that a G4 is faster at than a Pentium.

Apple is a speck on the PC world's radar.

I think you've got that backwards. The PC world is just a speck on Apple's radar. Apple is quite happy to go their own way and let the PC world go do... whatever it is that the PC world does. Every once in a while, the PC world takes a look at what Apple is doing and changes direction a bit, but that's about the limit of the interaction.

Re:Incredible!

[Anonymous Coward]

There's no primer, and just one coat.

http://www.djedwhite.com/photo/photo.php?dir=Def ec tive_Powerbook

Re:Incredible!

[afantee]

>> Only Mac users are gullible enough to APPLAUD a company making expensive Gigabit ethernet stock when so very few have networks capable of using its full potential.

I bet an idiot like you would prefer a floppy drive to Gigabit Ethernet and have never heard that you can just connect 2 Macs with a cable to make a Gigabit network.

>> And what's up with this 802.11g networking? Do you know how fast your cable modem is?

What the fuck you are talking about? A high speed wireless LAN doesn't depend on a cable modem.

>> And you wouldn't need the extra room for your apps if damn Apple would put back in the multiple terminals every other Unixy product in the world has. Damn Crippled Unix.

Have you ever used OS X or any other Unix, and what the hell are you on about? There are many terminal apps on OS X, you can run many instances with each having multiple windows, and even tabbed terminals.

>> And unfortunately, they probably WILL go under soon.

Apple is one of the few computer maker that are still profitable, and they have a $4,400,000,000 cash pile, so why don't you keep your stupid oppinion to yourself and go to play with your noisy and ugly PC toy.

>> They're getting squeezed to death because they don't run Windows. They're not compatible with Windows. They don't run Windows software.

Obviously you have never heard Virtual PC, which allows a Mac to run Linux, DOS and Windows 95/98/NT//2000/ME/XP, all at the same time if you have enough RAM. In addition, the best of main stream software from MS, Adobe, Macromedia, Sybase, Oracle, Bortland and FileMaker are available on Mac and usually work better than the PC versions. Don't forget that there are many Mac only softeare

Re:Incredible!

[drsmithy]

I know your joking, but I feel a /. beginning.

You'll actually find Steve Jobs started it, when he took a handful of carefully crafted Photoshop operations (most of which are only marginally faster) and extrapolated that to saying Macs are twice as fast and the so-called Mhz Myth.

Re:Incredible!

[GlassHeart]

Apple is quite happy to go their own way and let the PC world go do... whatever it is that the PC world does. Every once in a while, the PC world takes a look at what Apple is doing and changes direction a bit, but that's about the limit of the interaction.

Power Macs now ship with PCI, AGP, and IDE peripherals. Inside the box, I can think of MP3 and Samba as two obvious examples.

The point is not that the "PC world" invented any of these, but that Apple may not have made these design choices if not for their popularity in the "PC world".

So, no, Apple does look very closely at what happens on the other side.

Re:Incredible!

[Anonymous Coward]

+1, Insightful. Unpopular, but insightful.

Re:Incredible!

[Anonymous Coward]

I bet an idiot like you would prefer a floppy drive to Gigabit Ethernet and have never heard that you can just connect 2 Macs with a cable to make a Gigabit network.

A floppy drive is cheaper. Connecting 2 Macs is for kindergarten kids. That's not a network, that's a joke. If you guys had a clue, you'd be burning your crap to CD's and keeping them in an archive. Bang. "Network" no longer required. Except to look cool. No, I mean a REAL network. Go ask your boss (when you get out of high school) if he's got Gigabit ethernet. The answer's probably no. And if he does, he's too smart to let you put your home computer on it. So it's useless for the masses.

What the fuck you are talking about? A high speed wireless LAN doesn't depend on a cable modem.

Like I said, the majority of home users don't have a REAL network. They use their network cards for SURFING THE INTERNET WIRELESSLY. You've heard of Warchalking? Go figure it out. Humongous waste of a g card. That's not the bottleneck. It's the goddamn slowass bus and Ram. And Jeezus, rookie, stop swearing. You're scaring me.

Have you ever used OS X or any other Unix, and what the hell are you on about? There are many terminal apps on OS X, you can run many instances with each having multiple windows, and even tabbed terminals.

Sigh. Okay, retard, I'm talking Unix to you. You, being a Mac guy, aren't getting it. I'm talking about "virtual workspaces". Yes, I know about tabbed terminals, opacity settings, and all the other useless crap people are bloating their systems down with. I was just making the point that the Linux and BSD guys are giving away FREE operating systems that contain features that the expensive OSX Jaguar has purposely left out. Purposely. Get that. Because the features were in before Apple started OSX, then they were pulled out by Apple. Less features, more money. And that's a feature that I use daily. They think you're monkeys who can't handle multiple workspaces, or they're trying to sell monitors. Or they're afraid you'll really notice the lag that's not there on your PII 400Mhz running Gentoo.

Apple is one of the few computer maker that are still profitable, and they have a $4,400,000,000 cash pile, so why don't you keep your stupid oppinion to yourself and go to play with your noisy and ugly PC toy.

A, they're not very profitable right now. Look up their financial statements for the last year. B, I personally KNOW a guy who lives in my city with more than 4 Billion. ONE GUY. That's not a lot of money anymore. And Apple's bleeding money and marketshare. It will be 3 billion if they don't loosen up and become cool people again instead of DMCA-wielding, .Mac-overcharging, buggy iApp-releasing patent-threatening Cease-and-Desisting Lawyer scum. The old Apple was cool. The old Apple is dead. The new Apple is Microsoft with less marketshare. I liked Steve Jobs when he'd walk around the office with no shoes on. Now he wears suits, and *dresses down* for Keynotes because it's expected. They're way corporate inside and they've lost the spark of rebellion and hippieness. Now they're supporting DRM in Quicktime, as you can see from another Slashdot article today. They've gotta get "real" again or they're gonna die. Maybe they're dead already. At least if they're gonna die, they can die like men instead of corporate shills bending over for the MPAA and the Almighty Dollar. And C, you obviously don't have a Mac if you think PC's are loud. My PowerMac sounds like a swarm of angry locusts flying helicopters screaming into the microphone at a Guns'n'Roses concert.

Obviously you have never heard Virtual PC, which allows a Mac to run Linux, DOS and Windows 95/98/NT//2000/ME/XP, all at the same time if you have enough RAM. In addition, the best of main stream software from MS, Adobe, Macromedia, Sybase, Oracle, Bortland and FileMaker are available on Mac and usually work better than the PC versions. Don't forget that there are many Mac only softeare

Sorry. Virtual PC is a laggy bloated bastard compared to the real thing. Does as good a job as possible, but it can't compare with a non-emulated OS, even Windows. But that's irrelevant to my argument, which is that Microsoft has such a killer monopoly (especially now that Gates' golfing buddy George W. Bush made a *personal call* to the DoJ to have MS let off the hook), that Apple is bleeding to death because they are exactly what they have always longed to be: different. The common man doesn't want to think different, he wants what his neighbors have, and what he can get stolen software for easily. C'est la vie.

I hope Apple makes a turnaround in their own heads. I hope they truly embrace open-source. I hope they offer more customizability in their gear - including the ability to buy stripped down models of their gear. I hope they support the Linux guys. I hope they fire half their legal team - the bad half. I hope they defend their users from privacy invasion and stupid restrictions like DRM. I hope Apple never threatens the DMCA again. I hope they get cool with people hacking their stuff to improve it. I hope too much. None of this will happen. The die is cast, Apple is too corporate now. They will die soon. I'd start learning Linux commands now if I were you.

Re:Incredible!

[pi radians]

Hey thanks. Only one thing though...

You said:

Oh, wait, Macs are twice as slow when compared to Pentium IIIs with the same clock speed

So from the link you gave me I got this:

450 MHz G4 (2MB L2): 21.4/20.4

Well, the slowest pentium chip given on that page was:

650 MHz PIII (440BX): 31.3/22.4

So to find a slower (and equal clockspeed to the G4) I went to SPECbench.org and grabbed my own data:

450 MHz PIII (SE440BX2 Motherboard): 18.7/13.7

Now, I'm not some Mac zealot or anything. But it appears that you sir were mistaken.

Thanks for playing.

Re:Incredible!

[PCBman!]

*I* said nothing of the sort ;-). I think you meant the other guy.

Re:Incredible!

[pi radians]

My bad... Thanks for the link anyways. *slaps forehead*

Re:Incredible!

[Anonymous Coward]

I might not know all the famous buzzwords, or know that much about Unix and such, but I have a few comments to make and reply to.

"Go ask your boss (when you get out of high school) if he's got Gigabit ethernet. The answer's probably no. And if he does, he's too smart to let you put your home computer on it. So it's useless for the masses. "

If thats the reason, then why sell 3 Ghz P4's??? What person of the masses needs a 3 Ghz processor for word processing and emailing??? I know I use the Gigabit ethernet to its full when I bring my laptop into the office which uses a Gigabit switch.

"I was just making the point that the Linux and BSD guys are giving away FREE operating systems that contain features that the expensive OSX Jaguar has purposely left out. Purposely. Get that. Because the features were in before Apple started OSX, then they were pulled out by Apple. Less features, more money. And that's a feature that I use daily. They think you're monkeys who can't handle multiple workspaces, or they're trying to sell monitors. Or they're afraid you'll really notice the lag that's not there on your PII 400Mhz running Gentoo. "

Here, I have no idea what you are talking about...are you trying to say that Linux and BSD, the free versions, any Joe Schmoe can use right out of the box??? I highly doubt that. You are seriously underestimating a usable GUI. Last time I heard....getting a copy of Red Hat Linux, Yellow Lab Linux, or whatever CD wasnt free either.

"A, they're not very profitable right now. Look up their financial statements for the last year. B, I personally KNOW a guy who lives in my city with more than 4 Billion."

Actually, Apple IS profitable right now...what YOU arent seeing in the financial statements are the ONE TIME charges of the software companies they've been purchasing such as Shake. You take out all the purchases in the last year, and they made a profit. 4 Billion also isnt chump change for a company to have, especially a company the size of Apple. MS has about 32 Billion, yet is only about 100x larger than Apple. Bill Gates has more wealth personally than 95% of the companies in the US, but does that make them worthless, and they should shut down??

".Mac-overcharging, buggy iApp-releasing patent-threatening Cease-and-Desisting Lawyer scum."

Have you used .Mac?? They offer quite a bit of service that NO OTHER email company offers. If it was just an email address, I would agree, but its not, so hush. As for the iApps..what the hell are you talking about noob??? The iApps are the best apps for their market around (consumers)...you wont find a better MP3 player (iPod), you wont find a better Music player (iTunes), you wont find a better digital video editor (iMovie), you wont find a better DVD maker (iDVD), better photo processor (iPhoto), nor better calendar service (iCal) than Apple's iApps. Get a grip man.

"Maybe they're dead already. At least if they're gonna die, they can die like men instead of corporate shills bending over for the MPAA and the Almighty Dollar. And C, you obviously don't have a Mac if you think PC's are loud. My PowerMac sounds like a swarm of angry locusts flying helicopters screaming into the microphone at a Guns'n'Roses concert. "

They've been dead since 1990....that song is so old, cant you come up with a new one??? Apple and Corporate also goes together like cats and dogs. Apple has been anything BUT corporate with their products. Bottomline is Apple is a business just like every other business...and if having 4 billion in the bank, and having to make crucial business decisions because of shareholders, or selling products to make a buck make them corporate, then so be it. To compare them to MS only makes you look like a total fool though.

Bibble has been around forever on PC

[Anonymous Coward]

...and didn't anyone think about that fact that it could just be that he's become more efficient in the conversion procedure? Hell, it could just be that the Mac version is just written better. I have serious doubts about it truly being faster on a Mac (and it sucks that there's really no way to compare fairly).

Re:Bibble has been around forever on PC

[poisti]

>(and it sucks that there's really no way to compare fairly).
so you have to use real world apps to compare performance, in this case Bibble, or you could use Photoshop... curiously, Apple doesn't show many Photoshop benchmarks anymore and as other tests have shown, Intel pc's are faster at most things (videoediting, etc). However, this shows that the Altivec unit makes up for the increasing GHz/MHz difference between PowerPC chips and x86 chips

Re:Bibble has been around forever on PC

[poisti]

or I should rather say: it can under some circumstances partly make up for the speed difference

Re:Definitive Evidence

[Anonymous Coward]

Oh hell! Imagine if RIAA found Macs with n-speed CD burners in them...

"We found 5 dual CPU Macs with 20x CD burners - the equivalent of 170 PCs with CD burners, or 14000 ZX Spectrums with some kind of CD-write capability..."

who would have thought...

[BlowChunx]

that a multi-threaded app that utilized Altivec would beat a single thread that relied solely on the FPU to do the work...

I mean this is not rocket science! You would get similar results on most any machine using SSE2/MMX and hyper threading (perhaps...).

Re:who would have thought...

[jo_ham]

I think the point is that we had an article trashing the Mac for image processing because it was so slow at RAW processing. This appears to have fixed that problem, so there's no reason not to use a Mac in digital photography work now.

Re:who would have thought...

[Gropo]

You would get similar results on most any machine using SSE2/MMX and hyper threading (perhaps...)

What baffles me is that the distributed.net clients have never apparently taken advantage of x86 SIMD cores. You'd think that they would take advantage of whatever they could code for within these clients, as they far outnumber the MacOS clients, and the goal is to unlock an encrytion algorithm, not benchmark CPU's.

Rather than indicating that the distributed.net team would rather see PowerPC 74xx systems triumph in the key-crunching race, it would indicate that MMX/SSE2 are a royal pain in the ass to leverage unless you're coding/decoding pretty specifically what they were designed to code/decode - though IANAC++P...

Re:who would have thought...

[Anonymous Coward]

Depends. If they use GCC to compile it might not, but a good optimizing compiler such as Intel C++ or Visual Studio 2002/2003 will take advantage of SSE2/MMX in many routines.

Re:who would have thought...

[Anonymous Coward]

If you are thinking of the RC5 challenge, it is because there is no advantage to using MMX/SSE2 because there is no rotate left instruction (which is largely used only for the RC5 algorithm, looks like Motorola was targeting the PowerPC platform to being a SSL accelerated platform).

I'm sure the distributed.net developers are using enhanced instructions where they will be of use for practically any platform.

Re:who would have thought...

[Anonymous Coward]

Newer builds of GCC (3+) optimize at least as well as the Intel compiler. They also do so on many platforms, unlike the Intel compiler.

Re:who would have thought...

[Anonymous Coward]

so there's no reason not to use a Mac in digital photography work now.

Well, provided you happen to like the way Bibble does RAW conversions and you don't happen to use e.g. a Canon camera. This is just one program, with limited functionality, out of very many. All the others are still way slower on the Mac.

I'm not bashing Altivec, but you got to realize that specifically handtuning code will never be able to compete with a general CPU that is fast on everything. Altivec is great, most programs out there do not use it.

Re:who would have thought...

[Ninja Programmer]

• I think the point is that we had an article trashing the Mac for image processing because it was so slow at RAW processing.

Actually it was not just some article trashing the Mac, it was a very well thought out, controlled environment and fair study of the Mac versus PC performance.

Of course, what Bibble labs did in the privacy of their own facilities ... well its propaganda at best.

• This appears to have fixed that problem, so there's no reason not to use a Mac in digital photography work now.

No, the original author benchmarked 12 different scenarios in which Bibble's software was only used once. Only if Bibble is your primary software choice, does this matter at all.

Re:who would have thought...

[jo_ham]

True, but they're obviously going to try. Even if you don't have the goods you still try to sell your platform.

Re:Definitive Evidence

[Derwen]

MAY be faster? Do we have politicians submitting articles to /. now? An article completely devoid of a definitive answer... unbelieveable

Don't worry, by the time /. puts up a duplicate posting tomorrow, they may have a random opinion added :-/
- Derwen

Re:Isn't this pointless?

[bpbond]

No. See the many, many other discussion on /., ArsTechnica, etc., about G4 vector processing capabilities. This and laptops are the (only) areas where the G4 remains competitive or better than the P4.

and again /. fires off the flame/zealot war

[mousehouse]

argh... i absolutely loathe these mine's bigger and faster things... it's like a boy's pissing contest time-and-time again. this empty article blown up doesn't help! although i must add that it proves that decent programming skills _on_any_cpu_ helps build a fast program...

Biased...

[Justen]

The reality is that these "benchmarks" are, in all actuality, never really objective. The benchmarks from a few weeks ago were likely done by somebody who is less than a fan of the PowerPC G4 chip. The results from this article were written by someone who writes software for Windows and has decided to write a clean program for the G4 chip with its Altivec engine. Kudos to him.

The reality remains that benchmarks prove little.

People who are in love with Macintosh have, throughout history, had the speed card in their deck. At this particular time, many would argue they don't. (Many would argue they do...)

People whoa re in love with other platforms, hardware and software, like their platforms for specific reasons, as well. Speed may be one of them.

But, I think, deep down, Mac users are attached to the platform for more than just speed. It's the efficiency of the operating system, the attention to detail, the clean interface, the simple plug-and-play, the good support, the Apple iLife products...

It's all in the eye of the beholder.

jrbd

Re:Biased...

[milbybw]

Actually the original review was done by a Mac fan. While his benchmarks do not necessarily mean anything in a general sense (i.e. is a PC or Mac faster), they do have relevance in his specific situation. For the task of converting RAW images, his testing showed that PC's were faster. Many commented that it was due to poorly written software on the Mac, but that did not change the fact that if you wanted to do that task, you could do it faster on a PC. Now the situation may have changed (probably has). I would imagine that the original review will be updated with this new software for the Mac. Unfortunately it won't change the poor performace of reading the photos from the media...

Re:Biased...

[signer]

If you'd read the original article about the benchmarks from a few weeks ago, you'd know that the guy coming up with the tests was a Mac person. After using XP for a while, he wanted to know which one was faster at doing the tasks he routinely needed done. I absolutely agree about kudos to the MacBibble people and the reasons why people who stay with a Mac are attached to it. However, you should read the original article, and the comments responding to it (which include feedback from the original author clarifying some of his statements and expanding on his Windows and Mac experience) before assuming that negative comments are the work of people who dislike the platform/processor.

Re:Biased...

[Anonymous Coward]

Thank you for realizing the most important thing to most mac users, it's not the fact that they're fast machines, it's that they're GOOD machines.

Not only is the hardware of decent quality, but it runs all the software I want. I get the commercial packages such as office and cubase vst, and I can also drop to a terminal and apt-get basic unix packages, courtesy of fink.

I'm posting this from a dual 1ghz g4 with 15K SCSI drives. It's a fast machine, but more importantly, it's a good machine.

Re:Biased...

[davesag]

But, I think, deep down, Mac users are attached to the platform for more than just speed. It's the efficiency of the operating system, the attention to detail, the clean interface, the simple plug-and-play, the good support, the Apple iLife products...

haha unless you are a regular traveller. iCal sucks for travellers. try changing timezones. or scheduling flights. you'll miss your plane when iCal move your flight times for you to 'compensate' for the change in timezone.

Re:Biased...

[overunderunderdone]

The reality remains that benchmarks prove little.

I have to disagree. I would say that the particular benchmark in the original article and to a lesser extent even the much maligned Steve Jobs Photoshop benchmark are really the only kind that count - real world tasks that (some) users will actually be using (a lot). The problem is when people imply or believe that these narrow benchmarks of specific tasks mean something about the performance of the chips more generally. A VERY fair criticism of Steve Jobs, though to be fair his audience is largely made up of people that make their living with Photoshop. It's not fair at all of Rob Galbraith who's audience is entirely made up of people that make their living not only with photoshop but spend most of their time working with RAW formatted images.

The original article by Rob Galbraith was an extrememly fair test of how well different platfroms did a particular set of processor intensive/time consuming tasks that professional photographers are doing all day long. It doesn't matter WHY one performs better than the other. It doesn't matter if the test is "fair". It doesn't immediately matter* if the results are lousy on one because the software is written in a notoriously slow scripting language running under emulation and great on the other because it was written in assembly - it only matters that I can process more photos in an hour on one machine than I can on the other.

*One caveat to the above - in the long run it might matter that the tests are at least somewhat "fair" because the next upgrade to the software might change things dramatically - which apears to be the case with this second /. article.

Re:Biased...

[Kplusplus]

You should have sent feedback to Apple about this to Apple. iCal gets its time from the System Time and assumes your calendar is relative to that. The spec that it follows, has no support for relative timezones so In order to conpensate for them, YOU WOULD HAVE TO TELL APPLE TO ADD THIS.

I know that people may become engrained with software not working how it should since they accept so much piss poor software on the PC, that you merely accept it. On the mac things don't work the same, our software is better because we demand that it be, in vocal emails to program authors and providing feedback for things we find lacking in software.

Its more than just a computer, its an ideology about how things should be. That is why things are better on our side.

Re:Biased...

[Anonymous Coward]

You need to actually read the original article, and then follow the link therein to that site's discussion forum. The author is a digital photographer and long time mac user who had recent, first-time experience with a "Wintel" PC and decided to take a critical look at his platform of choice. I mention the discussion forum as the feedback he received, along with his thoroughly descriptive responses cut right through the kind of BS that inspires a post like yours.

The author is not a Windows programmer. He's a photographer. He acutally uses computers as a tool in his career. That's why you should read the forum for his extended discussion of how Windows XP was easier for him to use with his peripherals than OS X. We know you like Macs. I do too. But they aren't the Alpha and Omega of personal computers. As one of the author's more level-headed forum participants noted, it's dissapointing that the guy had to preface his remarks with a kind of "declaration of alegiance" to Apple products. Is it really so... verboten to point out that in some ways they might suck?

Re:Biased...

[davesag]

oh believe me I have raised hell about this both visa feedback and on the apple discussion fora.

Re:Biased...

[Anonymous Coward]

Actually, no it wasnt...the author of the original review has bashed Macs since day one, and has publicly stated his dislike for anything Apple. You only need check out the forums on DMN.

Hot Damn!

[The AtomicPunk]

Who gets to post the obligatory "people don't understand Mhz" post? :)

3DNow!

[turgid]

So, PCs have 3DNow!, SSE and SSE2 depending on what processor you have. I have observed factor-of-ten speed-ups of certain code using hand-crafted 3DNow! vs. GCC floating-point. I wonder how fast his algorithm would be if implemented in 3Dnow! or SSE? I bet my rusty old K6-2/500 could put in a reasonable showing at his benchmark.

Re:3DNow!

[caveat]

yeah, but IIRC AltiVec is a much cleaner, better implementation of a VPU than the x86 flavors (do they still share the FP registers a la MMX?) - so its code is still probably going to be faster than SSE optimized code (on a specialized black hole simulation that one of my former professors uses, i've seen a >20x speedup with good AltiVec code).

Re:3DNow!

[UberLame]

Apple provides a very nice, extremely easy to use Altivec library. It requires writing no assembly code, and I believe it even resorts back to non-Altivec means of execution if a program written using the library is executed on a G3. So, for instance, in Altivec, you can write things like:

result = vec_add( aVector, someOtherVector );

and it works properly regardless of what sort of vector you've chosen to use for aVector.

I've yet to see anything similar for 3D Now or SSE/SSE2. Everything I've seen for them is either a library that is too application specific (like a premade image recognition library), or requires using assembly and a compiler newer than VC++ 6.0 (maybe only SSE2 really requires that).

Apple also provides a bunch of other libraries, like vDSP (I'm sure AMD and Intel provide an equivelent), and BLAS (this is a somewhat standardized library across platforms. My recall is that there is a SSE/SSE2 version, but Intel charges money for it, instead of giving it out for free), and in general, they make it easier for Apple developers to take advantage of Altivec than Intel does SSE2 or AMD does 3D Now. Unfortuntaly, a lot of developers want to maintain only one code base across all platforms, so they won't use the Apple provided tools (there are free unoptimized versions of BLAS for every platform though, so developers should at least use that so they can't get speed benefit on platforms that provide it), which sucks because GCC also sucks for speed, so people using vendor supplied compilers on other platforms (like Intel's on Windows or Linux, SGI's on Irix, Sun's on Solaris) get a nice speed boost that would require hand assembly optimization to get on a G4.

Re:3DNow!

[be-fan]

Intel provides a nice, really easy to use C++ vector library as well. It provides C++ vector datatypes that make working with vectors the same as working with regular floats.

Re:3DNow!

[Lars T.]

Flashback!

Re:3DNow!

[andrewski]

FYI, there's no vDSP that I know of included with MS or Intel compilers.

Re:3DNow!

[Anonymous Coward]

I've yet to see anything similar for 3D Now or SSE/SSE2.

Check the documentation for the Intel C compiler or any recent version of gcc. Include "xmmintrin.h" and you will have wrappers for the vector instrinsic calls.

As somebody who has programmed both SSE and Altivec a lot I can tell you that it is really a myth that Altivec is so much easier - they are more or less identical. The main reason why most programmers aren't using SSE is that you usually don't need it with current x86 CPUs, while the PowerPC really crawls without altivec.

Re:3DNow!

[tshak]

Yes, AltiVec is nice, but do I HAVE to keep posting this link? John Carmack on the G4 and Altivec. (Note that this was before SSE2 was being considered)

Re:3DNow!

[Ninja Programmer]

Intel's compiler provides an even better solution -- don't do anything special in your source at all. Just flip the "compile with SSE or MMX" flag and your source is automatically vectorized.

Re:3DNow!

[UberLame]

You are, of course, correct. Intel does provide such a library. For a fee. Unless you can provide a link otherwise, I believe you are refering to Intel Performance Primitives ($199, not sure if royalty free, http://developer.intel.com/software/products/ipp/i pp30/) or Intel's Math Kernel Library, which is part of the Intel Performance Libraries (also $199, again not sure of royalty situation,http://developer.intel.com/software/prod ucts/perflib/).

I might consider purchasing these libraries if I really needed them, but in general, I find it unacceptable that Intel won't help developers get the most out of the platform for free. They already have some of the most expensive hardware in the business, and here they are charging us more.

Re:3DNow!

[UberLame]

No, not included. But Intel used to have a bunch of DSP code for free download from their web site. It used to be that you would find such things through their Code Central web site (http://cedar.intel.com/cgi-bin/ids.dll/topic.jsp? catCode=CAS). I'm not going to take the time to verify that such things are still there. And, it might not have been an exact replacement to vDSP. I was just trying to point out the ins and out of both sides.

Re:3DNow!

[UberLame]

From an assembly point of view, I believe that they are both about the same for ease of use. I think that initially Altivec had a bit of a lead, but not for long. I'm just arguing that the C libraries for using Altivec are better. And the file you pointed out just moves mmx/sse/sse2 programming a step up from inlined assembly. With this, I am still limited to the very small steps that inlined assembly provides. I don't see anything in this for doing dot products or cross products, although it would be trivial to write such functions, so perhaps I shouldn't complain so much.

BTW, I don't have it handy, so I'll just ask. Does MS Visual C++ 6.0 provide this stuff? I seem to recall that to use MSVC++ at all with SSE/SSE2 I had to either by the Intel Compiler or upgrade to VisualStudio.NET.

Hardly a fair comparison

[wiggys]

"Prelimenary tests indicate the Mac may be faster than PCs in RAW image conversion afterall."

Hang on a moment. The last Mac vs PC test was conducted fairly - Photoshop on a Mac vs Photoshop on a PC. Using nearly-identical software the clear answer was that the fastest PC today was faster than the fastest Mac.

Now someone writes more efficient code for the Mac, then tries to claim that Macs are somehow quicker than PCs? Talk about an unfair test - that's like that's like writing a pi calculator in BASIC for the PC and seeing how quickly it can calculate 1m decimal places on a 2ghz P4, then writing one in assembler for a Mac classic. If the Mac classic wins, does that mean the Mac* is faster at calculating pi than a PC?

* Macs in general

Re:Hardly a fair comparison

[Alien Being]

"fairly - Photoshop on a Mac vs Photoshop on a PC"

The name's the same, but we're talking about assembly level optimizations which necessarily differ between the two platforms. It's a different program for a different machine.

The fair test would be to pit the fastest Mac app against the fastest PC app.

Re:Hardly a fair comparison

[Nexx]

Well, that's one way of looking at it. Using Photoshop, a fairly well-written app on both platforms, one can measure the relative peaks on the architecture -- from the performance deltas, you could reasonably argue that the *very best* performance on the P4 will top that of the G4.

What the author is measuring, however, are the peak *real world* performance for the given task. Using your pi example, if the P4 variant is only available as written in BASIC, and the G4 variants are available as absolutely the best possible, and the G4 spanks the P4 *on that task alone*, it's still a meaningful comparison. In this (albeit contrived) case, *currently*, the G4 is faster than the P4 at calculating pi. Does that make sense?

Re:Hardly a fair comparison

[wiggys]

Yep, I know what you're getting at.

Re:Hardly a fair comparison

[kylner]

Actually, it's not entirely fair as it sounds. I could be wrong, but to my understanding Adobe is basically double Quartzing under OS X. What I mean by that is that Adobe coded in its own rendering display system for the Windows version of Photoshop to make it look slick, but never removed this rendering layer from the OS X version of it. That means the Mac OS X version has to both render it through Photoshop's screen rendering as well as Aqua. Talk about pulling double duty. Considering that it has to do that much extra work I think that the Mac hardware does an excellent job. It's Adobe's fault for not optimizing it under OS X. I don't know how and if that affects any raw image processing, but I thought it was a worthy knowledge nugget.

Re:Hardly a fair comparison

[overunderunderdone]

Hang on a moment. The last Mac vs PC test was conducted fairly - Photoshop on a Mac vs Photoshop on a PC.

No, you didn't read the article - it conducted tests using several different pieces of software but not one of them was Photoshop, although one was a third party photoshop plug-in. The tests were very narrowly focussed for a specific set of tasks of vital interest to professional digital photographers but of very little interest to anybody else.

The complaint at the time was that all of the software used was originally written for the PC and ported to the Mac. To use your analogy: In the original article the pi calculator was written in assembly for the PC but in basic for the Mac and the Mac suffered from that. Despite the MHz gap this was counterintuitive to those who follow this kind of issue because it was exactly the kind of specialized task where the PowerPC's superior vector/SIMD performance should have (and was assumed to) MORE than compensated for it's slower clock speed. Still it's a perfectly fair test because if you're interested in doing that task and this is the only software to do it with it doesn't matter to you WHY one system is faster than the other, only that it IS.

NOW, however as a follow up on the original story & controversy /. is pointing out that MacBibble has rewritten it's Mac version to take full advantage of the PowerPC's multithreading & Altivec processing and that it is much faster than it was before and therefore that one little task of interest only to professional photographers the Mac is back in the running as the fastest tool for that particular job.

A fair criticism of /. (though not of Rob Galbraith) would be that they are trying to imply that these very narrow and specific benchmarks are indicative of general processor performance or general vector performance when that is not the case. In the first story the implication was that the PowerPC's one advantage had ceased to exist when that was not necessarily the case. In the second story the implication was that the PowerPC's one advantage was a general one and not tied to very specific computing tasks. In slashdots defense - RTA, the /. crowd should be assumed to be technically inclined and able to pay attention to such details and to care about them.

Re:Hardly a fair comparison

[nelziq]

The point of the tests is to compare the platforms not just the hardware. It only makes sense to use the best software available for the platform. A power user would always use the best software, so the test should reflect the real world use.

Re:Hardly a fair comparison

[Anonymous Coward]

No - there's more to conducting tests than using the same software. The benchmark you're talking about used some heavy-duty filters that were optimized for x86 but not for the G4 processor. Not exactly a fair benchmark. It's *very* hard to do a fair benchmark across platforms.

Re:Hardly a fair comparison

[Anonymous Coward]

..case, *currently*, the G4 is faster than the P4 at calculating pi. Does that make sense?

That's a nice way of putting it. It really is faster at calculating pi. Unfortunately it is still way slower at calculating "e", "1/Pi", or "sqrt(Pi)".

My argument actually makes sense, because if you've downloaded Bibble you'll see that it only supports some Nikon models of RAW - all Canon users are fsck:ed, for example. So claiming that an optimized version of Bibble means the Mac isn't any slower for digital photography is plain wrong. If you happen to own a Nikon and are willing to shell out an extra $100, then one of the ten points where the Mac lost might not be so important. The remaining nine still stand.

Re:Hardly a fair comparison

[Anonymous Coward]

The point of the tests is to compare the platforms not just the hardware. It only makes sense to use the best software available for the platform. A power user would always use the best software, so the test should reflect the real world use.

Well, they did use the best software when they did the test. Or, do you say that they should have guessed that MacBibble might be improved? In that case, you should know that Eric is now working in the next release of PCBibble. Does that mean the current MacBibble results are invalid?

Finally, my digital SLR is a Canon (holding 75% of the professional market). These are not supported in Bibble, so for everybody but some nikon users these tests still represent the best possible programs.

Re:Hardly a fair comparison

[Anonymous Coward]

The complaint at the time was that all of the software used was originally written for the PC and ported to the Mac.

Sorry, but that's a non-argument. None of these programs used any hardware-specific stuff like SSE, but they are written in standard ANSI C and C++.

Now, I don't care where the programs were developed, but since 95% of the world aren't using Mac we can probably assume that most programs will not be developed on Macs.

If the Macs have huge performance problems with anything written in ANSI C/C++ that's a pretty big problem from my point of view.

Re:Hardly a fair comparison

[overunderunderdone]

Sorry, but that's a non-argument. None of these programs used any hardware-specific stuff like SSE

I've just read the original article and I didn't see this - could you give me a source? Frankly I would find it highly unlikely that a programmer wouldn't use hardware specific SIMD or vector processing since this is EXACTLY the kind of software that SIMD was invented for. If you are right these programmers should be fired - it is gross incompetance on their part to ignore a technology that would give them *massive* performance benefits. You could be right the second article/press release points out that MacBibble using hardware specific Altivec processing is faster than it's competitor by a factor of 10

...but since 95% of the world aren't using Mac we can probably assume that most programs will not be developed on Macs.

Yes, but 99.9% of the world are NOT professional photographers of that .1% that ARE probably 60-80% are using Macs. I'm being generous here, I have worked with many professional photographers and only one of the many I know uses a PC. Which is all beside the point, the complaint (I actually have no idea how legitimatly) was that the software was tweaked for the PC and the Mac suffered in comparison for that reason. The new Mac version is tweaked for the Mac and benefits *significantly* from that.

Re:Hardly a fair comparison

[dwightk]

Right, except Photoshop for PC and for Mac are not nearly-identical software. One is written for PC, one for Mac. Two separate sets of code sold under the same name

When will people realise...

[Nexum]

OK, I'm a techie and graphic designer (yes, rare).

When will people realise that raw speed, although useful to deisgners and artists, is NOT the be all and end all of which platform is preferable for this industry.

The main reason why macs are so dominant in publishing and art is becasue of the old (true) cliche - it just works. Designers are generally NOT a technical people, they think with the other side of their brain all day long, and technology confuses them, so even if a PC goes 20% faster at some filters, if they can't figure out problems with DLL's, conflicts, registry problems and having to reinstall Windows every 9 months then what is the better system for them?

How about usability and workflow (please comment on these only if you've used both machines (Win & OS X) in a demanding and very time specific industry to a large extent) - OS X hands down, allows me to ignore the fact that I am using very advanced technology that's incredibly advanced and *do my job*.

This allows me (and hundreds of thousands of others) to get a much bigger performance boost out of my work than a faster processor.

What are the productivity gains of perfect networking, great UI, better support for FireWire, BlueTooth, Wireless stuff etc etc etc.? It's not quantifiable but it is much more important than slightly faster processors, so lets just stop the whole thing there.

So in brief, processor speed important (and nice to see the Mac keeping up in one area) but not so important it outweighs the other thousand reasons design professionals use Macs.

-Nex

Re:When will people realise...

[wiggys]

The main reason why macs are so dominant in publishing and art is becasue of the old (true) cliche - it just works.

It's funny you should say that... the last 5 or 6 graphics PCs in our office have all come with Windows XP. We installed Photoshop, Illustrator, Pagemaker, hooked them up to our Samba server, network printers, scanners etc and... they just worked!

In fact, they have done for over a year now with virtually no problems at all.

And they have a right-mouse button and a scroll wheel!

Re:When will people realise...

[Nexum]

I think you should consider yourself lucky.

No offence, but I use a PC for programming, and my Mac for design work, I once tried to work solely on the PC for everything in between getting my new Mac, tried it for nine months, and did some traditional high end stuff. Speed aside, I found Win XP to be much less reliable to work with, I felt it was always trying to find a way to screw up behind my back - and I *do* think that your experience of XP is un-commonly good if you use the machines daily and hard.

But if it works, and you like it then good! (Although if you haven't used OS X (I'm guessing you switched from OS 9, or from no Macs at all) then I think you're missing out).

Oh, and search slashdot for 'Mac +Mouse' to read the 13'000 posts that describe how you can plug any 'normal' mouse into OS X and it'll work instantly...

-Nex

Re:When will people realise...

[wiggys]

I don't personally use the machines, I'm just the lowly IT admin who supports them. When we had Windows 95 and 98 machines I was kept very busy maintaining them, but since we've had XP our problems have been few and far between.

I can only speak for myself - I don't doubt that many people DO have nightmares with XP, I'm just saying that (so far) we've been lucky.

I guess I'm sceptical of the "Macs never crash and are easier to use" line because of my own (very limited experience) of OS 8.6

Re:When will people realise...

[Jeff DeMaagd]

Oh, and search slashdot for 'Mac +Mouse' to read the 13'000 posts that describe how you can plug any 'normal' mouse into OS X and it'll work instantly...

That part is actually the beauty of the USB spec.

If you design a device to the spec, then it will work on any OS that implements drivers to that spec. So you shouldn't need special Mac drivers for USB speakers, drives, mice, keyboards if they were properly designed. I even had a Mac friend that wasn't sure how to hook up his computer to his new reciever that has a USB jack. I told him to just plug it in. And it worked. The maker didn't even include a CD, drivers or any of that crap.

Also, the right to sell products with the USB logo means that the company is required to have those products compatibility tested - which means any product with the USB logo is going to work, so you don't have to look up special hardware compatibility lists. The reason why computer standards didn't work as well in the past is because those standards didn't mandate compatibility tests.

I know USB has its downsides but it works.

Re:When will people realise...

[splateagle]

mate, if the bulk of your Mac experience is limited to 8.6 do yourself a favour and find an excuse to play with OS X for a few weeks.

Before I bought my powerbook last year my only experience with Macs was classic (7.5.5 through 8.6) and I wouldn't have gone near 'em with a barge pole let alone blown two grand on a new Apple Laptop, an afternoon with OS X changed all that, and this was back before Jaguar.

seriously, give the modern Mac a real test-drive your inner geek will love you for it :)

Re:When will people realise...

[wiggys]

Don't you need drivers for USB graphics tablets, scanners, digital cameras etc on the Mac?

Re:When will people realise...

[rlsnyder]

There's also the whole critical mass thing. I work with a lot of graphic designers because, and they all use Macs because... they all know other graphic designers who all use Macs.. and on and on.

These guys - the guys I know, at least, are a close knit community. And technology just isn't that important to most of them. So they use a Mac because they know that if they get stuck, there's a whole host of other people trying to use the SAME exact software the SAME way on the SAME platform that will help them out.

Re:When will people realise...

[skribble]

Apparently the 'graphics' people in your office don't care about accurate color matching or anything like that. If so you'd need to add at least $500 worth of hardware and software and a few hours of clibration to come close to the accuracy a Mac has out of the box.

There's a hugh gap between "works" and "works well" and PC color matching incredibly sucks

BTW I use a three button scroll wheel mouse on my Mac just fine... that argument is as old, trite and wrong as can be... typical argument which just screams "I'm an Idiot"

Re:When will people realise...

[Anonymous Coward]

don't you still have to use "$500 worth of hardware and software and a few hours of clibration " to color calibrate your mac and printer?

Re:When will people realise...

[knoxer]

Well, I don't know about graphics tablets. But MOST scanners need some sort of driver. And MOST digital cameras do NOT need one.

Re:When will people realise...

[Anonymous Coward]

Comparing your OS 8.6 experience to OS 10.2.3 is about as fair as saying XP must be unreliable because Windows For Workgroups 3.11 was unreliable. Seriously, it's that different.

10.2.3 is basically just a unix box with a really nice GUI, extremely easy administration and commercial apps available.

And believe me, I'm no Mac fanboy. For years, I used a Mac solely because I had some sound software which crashed on NT but not on Mac. Since the release of 10.2, I use Mac because it kicks a whole lot of ass.

Re:When will people realise...

[haikvr]

i bought a cheap 50$ HP scanner from walmart - hooked it through USB and it worked - heck the HP manual did not have a word about OS X - and for windows you need to go through installing the crap and rebooting. Also i have hooked sony and canon digital cameras - works out fine - and i have also used the digital memory of the cameras to transfer files - as the camera memory shows up like a folder in the desktop - cant get any cooler !!!
get a mac :)

Re:When will people realise...

[liquidsin]

And they have a right-mouse button and a scroll wheel!

Yeah, so? The trackball I used to click 'Reply to This' has two left buttons, two right buttons, a scroll wheel, and six programmable buttons across the top (Kensington Turbo Mouse Pro). And it works just wonderfully on my Mac, as will any proper USB device. Oh yeah, and my computer colour matches properly, which I'm sure your PCs don't. And it may be a small thing, but I like my 'command' key on my Mac rather than 'ctrl+shift' on a PC.

Re:When will people realise...

[Anonymous Coward]

So they use a Mac because they know that if they get stuck, there's a whole host of other people trying to use the SAME exact software the SAME way on the SAME platform that will help them out.

Or how about: the company they bought it from can help them out. How many awards for technical support has Apple won now?

We've got a lab with several Dell PCs with Microsoft OS's, and it's a waste of time to try calling either Dell or Microsoft when something goes wrong. If you ask Dell, it's always a software problem. If you ask Microsoft, it's always a hardware configuration problem. Nobody takes any responsibility.

Re:When will people realise...

[Alan Partridge]

not really, the MacOS has a software colour calibration system called 'Colorsync' built in. It's MODERATELY sophisticated, but no match for a hardware calibrator - probably good enough for 70% of colour matching situations. Best part is that it's VERY integrated into the OS, and therefore widely supported by manufacturers.

It's really quite good.

Re:When will people realise...

[Steve Hamlin]

The main reason why macs are so dominant in publishing and art is becasue of the old (true) cliche - it just works...if they can't figure out problems with DLL's, conflicts, registry problems and having to reinstall Windows every 9 months then what is the better system for them?

What are the productivity gains of perfect networking, great UI, better support for FireWire, BlueTooth, Wireless stuff etc etc etc.? It's not quantifiable but it is much more important than slightly faster processors, so lets just stop the whole thing there.

----- -----

I agree with your premise - that in the end, the result is what matters, and if you can save 10 hours of headaches with sacrificing a few seconds here and there, then you are probably better off.

However, to your point above, please see Rob Galbraith's post about 10 down from the top of the discussion forum related to his comparison.

He states that he continues to use Macs as his primary machines. However:

• "In that vein, I think it's a mistake to let Apple continue to get away with saying their computers have an edge in the ease of setup and use department. Especially if you're a pro digital photographer.
• In my own experience of the past year is any indication, Apple's got nothing on Windows XP currently. Yes, you read that correctly, and I'm not currently on medication of any kind as I write this."

  "For a major project that ran through much of last year, I got up close and personal with Windows XP Professional running on the humble Dell box in the speed report. I connected a whole raft of pro digital SLR cameras, over a dozen card readers, plus several CD writers, several inkjet printers, a flatbed scanner and a film scanner. Every device connected and worked without a hitch, many of them sucking their own drivers from the ether and configuring themselves. Way, way cool."

  "On the Mac, it was as it always has been for me dealing with pro digital photography peripherals, whether in OS X or earlier iterations of the operating system. Some devices worked fine, though many required the manual installation of drivers, while some devices, and especially USB and FireWire card readers didn't work at all. Or required a driver for OS X 10.1, then a different one for 10.1.2, then a driver change again in OS X 10.1.3. Ugh. I've had fairly serious ongoing fights with my film scanner, so much so that I only use it on the PC now, where it just works. Where's the true plug and play in that?"

  "Part of this is just dumb luck of course, because with a different PC and different peripherals I could have been given a rougher ride by Windows XP, and an easier ride by the Mac. As it happens, however, life with Windows XP in 2002 was a breeze compared to the Mac. By OS X 10.2.3 things have settled down a lot on the Mac side, but for the speed report I experienced yet again an incompatibility between one card/reader combo that was not replicated on the two PCs. After awhile, these types of experiences make me think that Apple needs to spend more time delivering true plug and play for the pro digital photographer, and less time marketing the notion that they do."

  "Keep in mind, my preference would be to remain on the Mac, and right now, two of the key applications I use everyday are Mac only, so I'll boot up my Mac first every day for a while yet. But I won't stay on the Mac because of what I now consider to be outdated notions about the Macs ease of setup and use, since my experience using the other platform is that life is okay over there, even preferable in certain, specific ways."

Re:When will people realise...

[Jeff DeMaagd]

Don't you need drivers for USB graphics tablets, scanners, digital cameras etc on the Mac?

I think Apple has already integrated most such drivers into OSX, because the public interface to many common devices is specified by the USB standard. Any good USB digital camera would show up as a hard drive. Tablets probably fall under the existing Human Interface Device spec. Scanners would fit too. I don't know about printers though, I imagine having a specific driver might allow you to use more features.

It is even easy to buy several specific development demo boards from Cypress Semiconductor (among other companies) that show how you can design and produce 100% USB compliant keyboards, mass storage and other devices.

Re:When will people realise...

[afantee]

The fact that you are still talking about the age old single button mouse issue immediately disqualify you to give any informed oppinion about Mac. I have been using Unix, PC and Mac for over 10 years, and I can tell you that nothing comes close to Apple hardware and Mac OS X in terms of quality, style and usability.

Generally speaking, Mac users are passionate about their platform based on experiences, while most of the Wintel users often form strong oppinions out of ignorance. In most cases, Mac users win the argument because they are more articulate and intelligent than the average PC user.

I have noticed that there are lots of /. readers who have switched to Mac OS X (including several ./ editors like Rob Malda and Hemos) recently but very few Mac people switching to other platforms.

Re:When will people realise...

[Anonymous Coward]

I would use Macs, as most of my department do, but I find their design so incredibly ugly it's distracting. Plus I can't use .NET - when I can, I might consider it, and put the innards into a better looking box.

Re:When will people realise...

[hcdejong]

Right. He compares using Windows XP for a year to using eight versions of OS X over that same year. Well, duh. If you upgrade your OS, you've got to expect some changes. Windows has an advantage here in that XP is basically =NT4.2, a mature OS that has seen little change over the past few years.

Re:When will people realise...

[Anonymous Coward]

Its funny YOU should say Pagemaker.....Pagemaker has to be one of the WORST desktop publishing programs in the world, next to Microsoft Publish. No serious commercial print house/pre-press house uses it. I know because this is my field...Im not a programmer, but a production artist. People are forgetting the font management tools and color management tools on the Mac side as to why designers use Macs.

Re:Isn't this pointless?

[AssFace]

as I understand the PC has faster hardware in the sense that American cars have more horsepower - they just throw a ton of power at the problem and don't worry about the effeciency.
The Mac has the ability to do some cool wider pipline stuff and specialized vector processing - but you need to design stuff especially for it - otherwise it isn't as efficient and you lose to the big block Intel/AMD family.

I think the Playstation2 had this problem at first - it is *highly* optimized for vector processing and the first bit of releases for it hadn't taken full advantage of that.

If I can come up with a scenerio that is useful to me where I really *need* a mac, I'd consider it - but at this point, they are simply cute as hell and that is about it.

The speed of the hardware is irrelevant...

[Junior J. Junior III]

Unless the people coding the software take advantage of it. That's what I got out of the whole thing.

Anyone remember the Byte benchmark?

[wiggys]

If I recall correctly, Mac users used to trot this one out on a daily basis as "proof" that their Mac was faster than a Pentium III.

The reason it was faster was that the G3 had more on-chip cache, which suited the benchmark, and said absolutely NOTHING about the rest of the system.

A computer is as fast as its bottleneck... when evaluating performance it's best to see as many REAL WORLD benchmarks as possible. No use having a 12ghz processor if you still use 33mhz memory.

Re:Anyone remember the Byte benchmark?

[Anonymous Coward]

No, it was because the compiler for the G3 was able to realize some of the bit manipulations could be exploited by using the rotate instructions more effectively than the compiler for the x86 processor could. This led to a significant speedup in that particular portion of the benchmark.

Sager notebook (prev article reference)

[No-op]

I happen to own the Sager notebook that the previous article referenced (the alienware machine is a rebadged sager unit, at a higher cost.) I use it for a ton of RAW file conversions and it lays the smack down on my G4 mac, hands down.

I've completely stopped using the mac for all my conversion needs- maybe this app would be better, but really the speed difference is significant between the two platforms.

Maybe if I was willing to shell out $4k (USD) for a newer mac platform, just to get a few minutes faster at conversion, I could get some speed-up- but for that price I could buy two more of these laptops, with 2.8/3.06 Ghz procs and a gig of RAM. That's the typical Mac owner's conundrum. :)

Mind you, someone could write a SSE2 enabled RAW file converter, and it would perform the same way. hand crafted code that's optimized for speed using specialty CPU features is good for everyone, regardless of platform.

Now if only this guy could make CF cards transfer faster :)

Re:Sager notebook (prev article reference)

[splateagle]

take another look at the article: it doesn't talk about RAW conversions in general, but a specific optimised soloution which (by making best use of the stregths of the CPU in the Mac) runs rings around the PC at a quarter the number of cycles per second... There's no "maybe" about it, this App *is* better for handling RAW files, and it's better because it's running on a processor with appropriate stregths, but under normal un-optimised conditions you'll find that all the extra horsepower in your wintel box makes it that much faster.

under normal circumstances, if speed is your determining factor, it makes sense to go with a PC over a Mac until either the new IBM PPC chip matures for desktops, or Motarola extract the proverbial digit and make up the lost ground... either way (as others have already pointed out) for many of us raw speed isn't the determining factor...

Re:Sager notebook (prev article reference)

[Silverhammer]

Blockquoth the poster:

Maybe if I was willing to shell out $4k (USD) for a newer mac platform...

See, it's crap like this that tells me most Mac bashers really have no clue what they're talking about. The absolute supreme top-of-the-line PowerMac that was just introduced this past Tuesday only costs $3800. You have to add bloody RAID array to the configuration to even crack $4000...

...and that's only through Apple's own online store. Buy it anywhere else and you can get another $300 to $500 worth of free scanners, printers, cameras, and/or software.

Enough with the FUD already.

Re:Sager notebook (prev article reference)

[2nd Post!]

Correct me if I'm wrong...

but in the article, the Alienware was 2-3x faster than the 800MHz G4 on the Bibble software. Perhaps you don't have an 800MHz G4, perhaps it's only 400MHz (I don't think they were ever any slower), which means that your Saeger is possibly 4-6x faster than the slowest G4 I can imagine.

And in the previous article, it was shown that Bibble's software was about the same speed as the Nikon software, on the Alienware, and about twice as fast on the Mac, though the PC was still faster.

However, with the *new* MacBibble, the process is 10x faster than the native software, which means that if you have a 400MHz G4, MacBibble should be exactly the same speed as Bibble on your Alienware... and if you have anything faster, like 500MHz, or 550MHz, or 800MHz, or 867MHz, or whatever, your Mac (at least for RAW conversion), is now faster than your PC.

Why would you want to spend $4k when your current Mac might very well be faster than your current PC? What speed is your Mac, anyhow?

Re:Sager notebook (prev article reference)

[Anonymous Coward]

yeah
by all means $4000 dollars is nowhere near $3800.
He must have been mistaken in making that remark.

Re:Sager notebook (prev article reference)

[repetty]

Maybe if I was willing to shell out $4k (USD) for a newer mac platform...

You should ask your wife to take your checkbook away from you. You certainly don't know how to use it.

Re:Hardly a fair comparison - LIAR Mac faster in C

[Anonymous Coward]

Photoshop code is compiled at a DISADVANTAGE to the mac and Adode was caught taking large cache bribes to make certain Intel processors ruin faster than other CPUs. Adobe appologized saying it was an unfortunate sideeffect of "optimizations" PC people HATE the truth

The truth is... if source code is available and fair and not in Intel Assembnler, then running it on a mac is usually FASTER.

Macs are faster in most algorithms with source available.

Typically the PowerPC (seen in most of the the www.top500.org list of fastest clusters) trounces Intel and even AMD at almost every benchmark.

Not just the 10 famous benchmarks as part of the composite in ByteMark , but at many other things such as the RC5 contest.

according to the RC5 benchmarks AMD is far slower than dual cpu macintoshes (half as fast). (source available for cor rc5 loops for most
processors)

The Mac Dual 1 Ghz g4 is faster than all existing dual AMD motherboards in RC5 benchmark by almost 100%.

21,129,654 RC5 keyrate for dual 1 Ghz g4 system ! And Now apple sells dual 1.25 Ghz stock and this week a 1.45 Ghz which would be even faster.

A dual 1800+ AMD MP get only HALF as many as a Mac! 10,807,034 rc5 keys !

Funny "Mhz myth" there showing itself I guess... Apple now is selling even FASTER machines than that one I mentioned made over one year ago, but with smaller caches and less fast read-write ram (it
now uses DDR on newest boxes).

The mac I mentioned uses a 2 MB L3 cache and no amd mp dual cpu boards I know about have any L3 cache at all, so maybe that is why some common macs are
over twice as fast, its not just altivec meager tweaks to rc5. AMD have similar , but less mazing vector ops.

The Pentium 4 takes many cycles (over 7?) to do a simple left shift. That is why the Pentium is MUCH slower than even the AMD or Mac.

Most modern CPUs can do a left integer shift in 1 cycle, any barrel position, not 7 slow cycles.

(Shifting is used a lot in decryption, encryption, graphics processing and many things).

Another reason the mac might be over twice as fast as an amd dual mp board is not just the 2MB l3 cache but the fact that mac can read and write to
a cold page of memory simulatneously FASTER than any AMD MP designs which are biased for linear access and streaming. Many memory scatter
benchmarks show this too. Apples newest DDR-RAM machines might not offer this feature though.

True, RC5 fits in primary cache of most machines, though interrupt services need larger caches depending on interrupt designs and load for the rest of the OS.

The RC5 benchmarks are never run with interrupts off, they use real world overhead.

The Macs made since september also can RAPIDLY service every pci slot almost simultaneously one 32 byte cacheline each if needed. How can it do that ? Three cool features of modern PCI :

* out-of-order completion
* address bus streaming
* intervention

Out-of-order completion allows the memory controller to optimize the data bus efficiency by transferring whichever data is ready, rather than having to pass data across the bus in the order the transactions were posted on the bus. This means that a fast DDR SDRAM read can pass a slow PCI read, potentially enabling the processor to do more before it has to wait on the PCI data.

Address-bus streaming allows a single master on the bus to issue multiple address transactions back-to-back. This means that a single master can post addresses at the rate of one every two clocks, rather than one every three clocks, as it is in the 60x bus protocol.

Intervention is a cache-coherency optimization that improves performance for dual-processor systems. If one processor modifies some data, that data first gets stored only in that processor's cache. If the other processor then wants that data, it needs to get the new modified values. In previous systems, the first processor must write the modified data to memory and then the second processor can read the correct values from memory. With intervention, the first processor sends the data directly to the second processor, reducing latency by a factor of ten or more.

ALtivec is not usually the reason a mac performs better than Intel in benchmarks of properly compiled code, because the famous set of 10 algorithms in ByteMark were not using ANY altivec instructions.

And the AMD bests the Intel at Rc5 mainly from integer features.

I laugh when pc people try to dismiss the fastest machine (Macs) by claiming Altivec "cheating" all the time. The mac people should be the ones to call foul when Intel was cuaght PAYING adobe to slow down filters in one version of Photoshop to artificially make the Pentium MMX 166 Mhz look faster. They got caught paying big bucks. Adobe replied that it was an unfortunate side effect of adding optimization for MMX and not keeping the code efficient in the non MMX case as it was before. HA!

Almost every pc person likes to use benchmarks that use lots of assembly for intel (Quake, etc), but shy away from benchmarks that offer source code in ANSI C.

I knew the mac handled RAW better than PCs and this news is no surprise to me.

Using photoshop while ignoring the famous msaller area filters and concentrating on large IO intensive filters is obscene.

Re:So what?

[splateagle]

same could be said of any /. post that doesn't match your specific interest set.

point is that it's tech related and of interest to plenty of /.ers so that's why it's here, if you're not interested I suggest you read something else.

Re:The mac is fastest at RC5 and tons of routines.

[MonTemplar]

-1, Pointless Willy-waving.

Slashdot Slashdotted?

[Anonymous Coward]

Lateley it's been slow and sometimes won't even respond. It's like me, kinda.

Re:Slashdot Slashdotted?

[teamhasnoi]

Haven't you heard? It's running on a Mac.

/me ducks and runs

posted from Chimera ;)

Re:Slashdot Slashdotted?

[PetWolverine]

I thought they must just be using MSSQL on Windows.

But then I remembered that this is /., and they know better than to do that.

vague test results

[Anonymous Coward]

this test is vague at best.
Run the macbibble algorithm on a SINGLE processor PC and see what the time is then.
Dont compare it to NIKON Capture 3.5 which could be doing a additional things during the conversion process depending on what algorithm was used.

Re:Hardly a fair comparison - LIAR Mac faster in C

[wiggys]

Adode was caught taking large cache bribes

cache bribes? How much did Intel give them, 64K or more?

Re:Isn't this pointless?

[Anonymous Coward]

Because it undoubtedly makes use of the G4's AltiVec

Apple propaganda here. They call AltiVec "The Velocity Engine" though, in a continued display of marketing department stupidity.

This is Ridiculous !!

[Anonymous Coward]

Come on People. Wake up.
The older article compared results between the same software (whether it was Nikon, Photoshop, or whatever) on Different OS's.
This article is comparing 2 different Software programs on 2 different OS's. This IS NOT A FAIR COMPARISON.
Can even the apple zealots that have joined ./ please admit this much.
Are we gonna compare a porsche's 0-60 on snow with a geo metro's 0-60 time on road ???

Re:This is Ridiculous !!

[Anonymous Coward]

oops. please excuse me for not rereading my reply before posting it.
I meant to say that the older article compared numerous programs between both operating systems. Whereas this one just runs 2 different software results on OSX. The conclusions that is being drawn without showing how these 2 programs work on a PC are weak. This kind of test would be laughable in the engineering or scientific community. The older article had thorough tests between 4 different computer setups.
This is just proof that mac enthusiasts will come up with any unfair comparison to give themself reason to beat their drums louder.

Re:Isn't this pointless?

[Anonymous Coward]

Won't this claim be pointless once someone copies this algorithm

The Mac program makes use of hardware only available on the mac.

jeezopetes

[bdowne01]

I'm really getting tired of the whole Mac vs. PC war being based on speed.

I'm not really sure how many times it has to be said, but a great number of Mac users don't use Macs because they're faster. In fact, let me say it again:

It's not about speed

I really can't believe that with the Slashdot community--being so "in tune" with corporate ploys and runaway marketing tactics--still fall for the MHz propaganda, and the speed benchmarks that accompany it.

Since when is the most important thing about a computer the speed? Granted, if you're playing BitchBlaster 2023 that requires a GeForce9000 Mx2+3.144 video card, maybe.

But I'm not sure if people noticed: Most Mac people aren't die-hard gamers. Macs aren't great gaming platforms anyway. They're for people that do work with their computers and rely on them.

These people care not about the absolute speed of their Mac, rather, they care that it works every time that it is booted and that the end-user experience is much more pleasant than someone using something like Windows XP.

So please, people of Slashdot--I know you have above average intelligence:

It's not about speed.

Re:jeezopetes

[desolation angel]

So please, people of Slashdot--I know you have above average intelligence:

You really haven't been here long have you.

Re:jeezopetes

[alkini]

I agree, for the most part.

For many Mac fans it's about the speed in terms of entire tasks, not simply the computing time.

If, for example, it takes you 5 minutes to get your PC to do what you want it to, and then it does the task in virtually no time, what does that matter if you can do the same task on a Mac in 30 seconds and wait an entire second for it to complete the task?

Obviously the numbers are just there to illustrate my point, not based on anything specific.

I guess what I'm saying is that it's about speed including usability and reliability.

Re:jeezopetes

[Anonymous Coward]

It's been a couple of weeks since I read the original article, however I do remember the author indicated that his new version of XP rivalled the Mac on ease of use and setup. The hardware he added simply 'worked' when he plugged it in.

You are right, it's not about speed, and the article touched on that point as well.

Re:jeezopetes

[lasmith05]

The reason most mac people aren't hardcore gamers is because for the longest time, hardcore games were NOT avaialble for the mac. The best FPS the mac had for a long time was marathon, and IMHO it sucked compared to doom/quake/etc. If a game did come out for the mac, it usually came out 1-3 years after the PC release, and by then it would be old news. Now 5-10 years later, no respectable computer gamer would even think of getting a mac for gaming, even though more games released now work on both mac and pc. The PC has a huge headstart.

Re:jeezopetes

[Anonymous Coward]

Hardcore gamers...very true about being PC instead of Mac....but why buy a $2,000 USD PC just for games then?? WHy not spend $200 on a Playstation 2 and have more games to boot??? Dont get me wrong, I know all about the higher resolutions on PC's, as well as multiplayer experiences...but still....thats alot for a gaming machine. In my situation, I have a PC next to my 3 Macs not only for games, but for certain 3D apps that arent Mac yet. If it werent for those few holdovers, this PC would be out the windows...pun intended. LOL

Re:Isn't this pointless?

[AssFace]

no no no - you got it all wrong - I'm stringing letters together. not entire words. the words are simply an epiphenomena that you see as a result of me flailing away against my keyboard.

I just close my eyes and bang on the keys and then as a result, you see this.

Re:Isn't this pointless?

[Anonymous Coward]

I write c++ code for both pentium 3/4 and the g4. For most conventional operations, the pentium is going to smoke the g4. Except if you have a data parallel problem. I wrote a vectorized monte carlo integration routine that was 2.5x faster on my g4 laptop (800mhz) than on my p4 desktop (1700 mhz). I didn't use SSE1/SSE2 on the P4, using altivec extensions is much easier than SSE2 extensions; since there is a C language interface and no context switching going on. What does this have to do with pictures? Photographic data is a prime example of data that can be vectorized during manipulation.

If I could only afford it !

[Anonymous Coward]

I know this is little offtopic but I want Mac so badly. Tried Xandros, dont' have time for Gentoo to finish compiling and probably give Mandrake 9.1 a try.
I lived in the darkness for so many months not knowing how many good applications for Mac are out there. You can even run Virtual PC if there is a need for Window$. Thanks guys at CompUSA.
Anyway I've read that article couple days ago about summer MacExpo. Will the prices on G4's drop after the show and where is a good place to look for pre-owned G4's ?

Re:If I could only afford it !

[splateagle]

if you're on the level, LowEndMac is a great resource for this, they've got bag-loads of content on running older systems and pointers/links for where to find good deals on used systems. happy hunting!

It's an OS issue...

[salimma]

... until recently the most common bundled OSes with PCs, the Windows 9x series are based on DOS, and thus does not support SMP.

Even today the only way to get a branded SMP box is if you splash out a *lot* of money on a 'workstation' model. Most people just build their own box.

Funny how a lot of these same users bashed MacOS (up to 9.x) for not having true preemptive multitasking, etc. etc. ...

I am still having problems even now with SMP under Windows... the driver for my Handspring Tréo keeps crashing :|

Re:It's an OS issue...

[salimma]

Funny how you are bashing windows for crashing when the problem lies with drivers from a 3rd party driver from Handspring

The two are related; the relatively few Windows SMP boxes lead to most developers only testing on single-CPU systems. And the lack of consumer-grade SMP-capable OSes for a long time (even now Windows XP Home does not support SMP) partially contributes to this.

Contrast to BeOS which debuted as a dual-CPU computer. The accepted norm is to develop multi-threaded apps by default, not as an afterthought.

Oh, and I am not a Mac troll. An anti-Windows troll maybe, and a pro-Unix troll probably, but are you not like a pot calling a kettle black? At least I don't declare in public my obsession with young actresses :p

....SO!?

[Anonymous Coward]

so what if the Mac is faster in RAW whatever !? ....my penis is larger !!!!

Re:....SO!?

[Anonymous Coward]

my penis is larger ... than? without a comparison this assertion is meaningless, and since I suspect an appropriate comparison would be "a paperclip" it's hardly something to be wasting all those exclamation points on now is it?

Re:....SO!?

[Anonymous Coward]

It's so funny when people respond to such things. Thanks for a good laugh! The ./ crowd is so predictable some(all)times.

It's All In The Interpretation

[Dr. Wu]

I honestly don't pay much attention to side-by-side comparisons, unless the systems themselves are significantly similar. To me, comparing an Apple to a PC is akin to doing a comparison between an Xbox and a PS2. Both systems will outperform the other when using certain tests, while in other cases they will be similar.

It all comes down to a combination of hardware and software, and it's relatively easy to skew the results either way using these factors. So getting an unbiased test is going to be very unlikely, even in the best of conditions.

My motto is, if it works for you, go with it.

Dr. Wu

Macintosh seems to excell at most ANSI C benchmark

[Anonymous Coward]

Sorry for having to post this again. An AMD fanboy censored and modded down my first post of it as redundant my http://apple.slashdot.org/comments.pl?sid=52307&ci d=5188424
even though it was not redundant in ANY way at Thursday January 30, @08:38AM when it was posted. Therefore I had to resubmit the following:

Most pro mac posts get -1 on slashdot as all of my INFORMATIVE posts do.

Hurray to Bibble Labs! But Hurray for ANSI C too, not just Altivec. The Mac has regularly been faster even when not a single altivec (SIMD) instruction was used. Besides Macs Altivec is not cheating because its a function of the c compiler design NOT ASSEMBLER usually. Unlike MMX, 3dNow , etc.

==========

Typically the PowerPC (seen in most of the the www.top500.org list of fastest clusters) trounces Intel and even AMD at almost every benchmark.

Not just the 10 famous benchmarks as part of the composite in ByteMark , but at many other things such as the RC5 contest.

according to the RC5 benchmarks AMD is far slower than dual cpu macintoshes (half as fast). (source available for cor rc5 loops for most
processors)

The Mac Dual 1 Ghz g4 is faster than all existing dual AMD motherboards in RC5 benchmark by almost 100%.

21,129,654 RC5 keyrate for dual 1 Ghz g4 system ! And Now apple sells dual 1.25 Ghz stock and this week a 1.45 Ghz which would be even faster.

A dual 1800+ AMD MP get only HALF as many as a Mac! 10,807,034 rc5 keys !

Funny "Mhz myth" there showing itself I guess... Apple now is selling even FASTER machines than that one I mentioned made over one year ago, but with smaller caches and less fast read-write ram (it
now uses DDR on newest boxes).

The mac I mentioned uses a 2 MB L3 cache and no amd mp dual cpu boards I know about have any L3 cache at all, so maybe that is why some common macs are
over twice as fast, its not just altivec meager tweaks to rc5. AMD have similar , but less mazing vector ops.

The Pentium 4 takes many cycles (over 7?) to do a simple left shift. That is why the Pentium is MUCH slower than even the AMD or Mac.

Most modern CPUs can do a left integer shift in 1 cycle, any barrel position, not 7 slow cycles.

(Shifting is used a lot in decryption, encryption, graphics processing and many things).

Another reason the mac might be over twice as fast as an amd dual mp board is not just the 2MB l3 cache but the fact that mac can read and write to
a cold page of memory simulatneously FASTER than any AMD MP designs which are biased for linear access and streaming. Many memory scatter
benchmarks show this too. Apples newest DDR-RAM machines might not offer this feature though.

True, RC5 fits in primary cache of most machines, though interrupt services need larger caches depending on interrupt designs and load for the rest of the OS.

The RC5 benchmarks are never run with interrupts off, they use real world overhead.

The Macs made since september also can RAPIDLY service every pci slot almost simultaneously one 32 byte cacheline each if needed. How can it do that ? Three cool features of modern mem conttroller bridge to PCI :

* out-of-order completion
* address bus streaming
* intervention

Out-of-order completion allows the memory controller to optimize the data bus efficiency by transferring whichever data is ready, rather than having to pass data across the bus in the order the transactions were posted on the bus. This means that a fast DDR SDRAM read can pass a slow PCI read, potentially enabling the processor to do more before it has to wait on the PCI data.

Address-bus streaming allows a single master on the bus to issue multiple address transactions back-to-back. This means that a single master can post addresses at the rate of one every two clocks, rather than one every three clocks, as it is in the 60x bus protocol.

Intervention is a cache-coherency optimization that improves performance for dual-processor systems. If one processor modifies some data, that data first gets stored only in that processor's cache. If the other processor then wants that data, it needs to get the new modified values. In previous systems, the first processor must write the modified data to memory and then the second processor can read the correct values from memory. With intervention, the first processor sends the data directly to the second processor, reducing latency by a factor of ten or more.

ALtivec is not usually the reason a mac performs better than Intel in benchmarks of properly compiled code, because the famous set of 10 algorithms in ByteMark were not using ANY altivec instructions.

And the AMD bests the Intel at Rc5 mainly from integer features.

I laugh when pc people try to dismiss the fastest machine (Macs) by claiming Altivec "cheating" all the time. The mac people should be the ones to call foul when Intel was cuaght PAYING adobe to slow down filters in one version of Photoshop to artificially make the Pentium MMX 166 Mhz look faster. They got caught paying big bucks. Adobe replied that it was an unfortunate side effect of adding optimization for MMX and not keeping the code efficient in the non MMX case as it was before. HA!

Almost every pc person likes to use benchmarks that use lots of assembly for intel (Quake, etc), but shy away from benchmarks that offer source code in ANSI C.

I knew the mac handled RAW better than PCs and this news is no surprise to me.

again sorry for reposting, its just that one fanboy is modding my stuff as redundant and I had to repost to make my words known even though comments.pl?sid=52307&cid=5188424 was first in all issues coverred and not redundanty in any way based on its timestamp.

Adobe Photoshop 7.0.x AltiVecCore Update plug-in

[tholomyes]

If you've read this far you might be interested to note this plug-in from Adobe that "enhances the reliability of Adobe® Photoshop® 7.0.x software running on a Mac OS X system that uses the G4 processor" from a couple of days ago.

No word on whether this gives the PS on G4s any kind of speed boost, though.

Re:Isn't this pointless?

[Anonymous Coward]

No because the algorithm seems to utilize the altivec instruction set which isn't present on processors PCs use

Perfect Example between Mac vs. PC

[Anonymous Coward]

PC Favoring Article:
Tests done of 4 computers, 4 different
processors, 2 OS's,
Over 30 tasks done using 6 different Programs.

Mac Favoring Article:
Tests done on 1 computer, 1 OS, 1 Processor,
1 Task done on 2 different Programs.

It just appears to me that this is an unfair comparison. It seems that the conclusions of the former test are founded on principals of scientfic testing and have more credibility. Whereas the conclusions of the latter article are amusing at best.

Re:Perfect Example between Mac vs. PC

[caveat]

It seems that the conclusions of the former test are founded on principals of scientfic testing and have more credibility

i am a scientist, and neither of these tests has any sort of scientificism about them. you got the second one ok, but the first one used a poor choice of software (should have been the leanest, most optimized package for each system), a very limited set of test processes, and incredibly disparate hardware - yes, it was supposed to be 'top of the line' systems, but it's really bananas to cherries here; it's like dragging an extremely elegant, sophisticated, state-of-the-art 800hp F1 car up against an old-school brute-force 8000hp Top Fuel dragster and declaring the dragster a much better overall car, because it can romp in the 1/4. by that reasoning, i could say that this rock repels tigers...

Re:Isn't this pointless?

[Anonymous Coward]

Thanks for the clarification. So few people bother with this anymore, but I appreciate it.

Re:New phrase coined:

[Anonymous Coward]

LOL you know I've got an old Powerbook "Lombard" and until I upgraded the processor to a G4 I got to watch a lot of the beachball too... It seems that Apple has done a fair amount of Altivec optimizations theirselves in Jaguar.

the original test

[catwh0re]

the original test that found pc's to be more graphically applicable was rather contrived anyway, and this was before taking into account the tests were completed on recent pc hardware, and then against 6+ month old apple hardware.

Now that the powermac's were just updated, it would be interesting to see how the results would differ.

(I argue that the original tasks were particularly x86 friendly with focus on sse etc, and then no focus on the comparable altivec, basically a set of tasks chosen that would favour PC's all along, and not accurately reflect graphic designers actual work habits.)

There's that matter of cost....

[Mulletproof]

Not only does ineffecient method you described work, but it's generally cheaper than going the Mac route. Under those circumstances, I'll take the hit in effecientcy.

Re:There's that matter of cost....

[AssFace]

That's how I see it as well.

For the things that I do, I can't justify anything beyond AMD and linux (or freebsd).

but I do like the shiny laptops - at least the way they look :)

Re:New phrase coined:

[mbbac]

The beachball is displayed if the current window is not responding. You can mouse over the window of any other program (or window in another thread) and it'll disapear and you can get to work with that one while the other window chugs away at whatever it was doing (or in some cases until you force quit the application).

So, it's not a Beachball of Death. Maybe it is a Beachball of Partial Death.

Sweet Mother of God

[wazzzup]

Will this never end?

I love Macs, I've used them exclusively for over 10 years now and don't see myself switching anytime soon. Given that...

To Mac zealots:
PC are faster than Macs. Get over it. Yes the PPC chip is more elegant and efficient but it runs slow (relative to Intel). Good Altivec applications are few and far between and don't really apply to the day-to-day home and business user. If the PPC 970 comes out this summer, then maybe Macs will again TEMPORARILY hold the speed crown but until then, PC are faster by using brute force. If sheer computing performance is your #1 requirement, then a PC should be your choice. If you're poor and only have $400 to make sure your child has a computer, then a PC is your only choice. Don't even start by saying with that money you could buy some 1997 era Mac either. Please.

To PC zealots:
The overall user experience on an OS X system outweighs the fact that Win XP may idle faster when running Word. In those applications that can take advantage of vector processing, Altivec is far superior to 3DNow and SSE. Plus, I see a lot of complaining about the program was written explicitly for the Mac so the comparison is unfair. Welcome to our world. Most software written to support hardware (scanners, cameras, etc.) is a blatant PC port of a hastily written "good enough" POS program. Plus, Mac laptops have better battery life AND get the full desktop chip, not some crippled "mobile" version designed to prevent penile burns and 20 minute battery life.

Personally, I'll take elegant and efficient any day. Quite frankly, I'm glad the PPC has temporarily lagged behind. It's forced Apple to really tighten up things to keep competitive and it shows. This might not have happened if the processor would make up for any code bloat and inefficiency. Look at Safari - 3MB download. Look at OS X speed from 10.0 to 10.2. Phenomenal. When the 970 comes around, OS X should theoretically run like a champ.

Re:Sweet Mother of God

[PetWolverine]

I have one bone to pick:

Don't even start by saying with that money you could buy some 1997 era Mac either.

The $400 Wintel will be using essentially five-year-old hardware anyway. Celerons just suck, even compared with the G3. Maybe 1997 is a little extreme, but if you have money to get an iMac from 2000 or so, so that you can run OS X, I'd say that's preferable to some $400 POS stumbling (as opposed to running) XP.

Other than that, point taken, including the fact that Wintels are faster than Macs.

The thing about this is that whichever one is on top at a given time is only on top temporarily--they've traded back and forth countless times in the <20 year history of the platforms. When the Mac came out, I'm sure it was faster than any DOS crap offered at the time; by the time of the 68040, it was lagging behind; the PPC brought us back into the lead; when the 604e was aging the Pentiums of the time (P2s, I believe) were faster; the G3 turned the tables back in our favor dramatically; the G3 started to age, the G4 came out, now the G4 is aging and the 970 is almost ready to replace it.

This is the real reason for the fact cited numerous times in the comments above that speed isn't really an issue. If you're selecting a platform to stick with for the long term, speed should be the last thing on your mind.

Re:Sweet Mother of God

[geekee]

Hmm. I wonder which zealot you are?

Re:Sweet Mother of God

[Anonymous Coward]

not some crippled "mobile" version designed to prevent penile burns

Are you implying that penile burn is a good thing?

Re:Sweet Mother of God

[CTho9305]

What the heck is with people saying PPC is "elegant" and x86 is "brute force"? The P4 has an EXTREMELY elegant core to execute instructions so fast. What makes PPC elegant? A higher IPC? Is it really that elegant if it costs more to do the same thing?

I use a PIC for some work. Its IPC is 1 - every instruction (except branches) take exactly one clock cycle. Is this PIC elegant? Maybe, by your definition. Is it a good solution for a desktop because it of its IPC? No way! It is 20 MHz.

I just don't get it. I think the whole "elegant" vs. "brute force" claim is just so Mac fanboys can feel good about something they really don't undersatnd.

Who Cares?

[TheRaven64]

I have a 1.33GHz Athlon. I have a CPU usage graph sitting in my system tray. My CPU usage almost never goes above 20% (exceptions: Compiling and encoding oggs, which will use 100% CPU however fast your CPU is). On a new Mac, a lot of the GUI related CPU load is shunted to the GPU, and PPC chips do run faster than x86 chips per MHz (This was never in dispute. The dispute is that a 1GHz PPC can outperform a 3GHz x86, which stretches even my 'will-to-believe'). So, If I upgrade to a new Mac with Dual 1.42GHz CPUs I get

1. A faster CPU
2. A spare CPU
3. A much nicer OS (I currently dual boot Win2K and FreeBSD. Win2k is less hastle than BSD, BSD is more powerful, but requires occasional tweaking.)
4. A better all-round system. The CPU is not everything. Firewire 800, 802.11g, Bluetooth etc all add to the system
5. A computer that's been designed, rather than agregated, as the PC was.

And the reason I'm still using a PC? Cost. At the moment, my 18-month old system really isn't slow enough to justify upgrading it.

Re:Who Cares?

[Alan Partridge]

you won't get your 'spare' CPU I'm afraid, MacOSX does a LOT of load balancing to remain responsive under load - and in THAT area it succeeds handsomely. Certainly, MacOSX 10.2.3 is by far the best multitasking OS I've used - even on a 450dualie you can run tens of demanding apps while listening to music skip-free or attending to email without noticeable effects.

I just wish Flash player performance was better!

Re:Who Cares?

[TheRaven64]

So I only get 2 spare half CPUs? Oh well...

Re:Who Cares?

[Anonymous Coward]

"Certainly, MacOSX 10.2.3 is by far the best multitasking OS I've used"

You never used BeOS then, did you?

This might be an oversimplification, but . . .

[Anonymous Coward]

Wouldn't it be logical to assume if we need convoluted and contrived benchmarks to decide which machine is faster, it might just be that they run *about* the same speed? If our perception doesn't allow us to distinguish easily which one is faster, then what is the difference if xyz Benchmark gives the Mac .5% faster ratings?

Oh well, that's the way I see it.

Go home folks, nothing to see here...

[Anonymous Coward]

For those who didn't know, MacBibble was already like 3 times faster than Nikon's Mac software. Making it multi-threaded makes it twice as fast. The latest AltiVec improvements only increased the speed 50-70%.

If Rob Galbraith ran the same benchmarks with a dual-proc P4 and the new MacBibble on a dual-proc Mac, the PC would still win, just by a smaller margin.

Re:Go home folks, nothing to see here...

[Alan Partridge]

dual proc P4?

who makes one of those?

If you mean Xeon, then you're paying EVEN MORE than Apple charge; if you mean dual Athlon - you'll be building it yourself! (not that it's a bad idea...)

Re:Go home folks, nothing to see here...

[Anonymous Coward]

The price difference between Xeons and P4s of the same speed is pretty minimal on Pricewatch. Let's see...

$300 for a dual-Xeon MB (onboard AV/Enet)
$900 for 2x 2.8GHz Xeons
$400 for 2GB of DDR
$300 for 200GB 7200rpm HD
$100 for a case
$200 for a DVD/RW drive
$200 for Win XP Pro

That comes to $2400. Am I missing anything?

The Mac Rob Galbraith used is a Dual 1.25GHz with 1GB RAM and a 120GB HD. He estimated its price at $3600 (now $2600), which right now would buy you a dual 1.42 with 2GB RAM and 180GB HD.

It's quite likely that the dual 1.42GHz machine will be faster at Rob's benchmarks than the 1.25GHz one, but not by much because the bus isn't any faster. Keep in mind that Rob used a Beta version of MacBibble that already included multithreading and some G4 optimization (though maybe not as much as in the release version).

This is why the original benchmark was good

[Anonymous Coward]

Both systems were running the same software on the mac and the PC. The PC won most tests. It may be that the PC has better compilers or other advantages but the point is the PC won using the same software.

Speaking of gaming consoles another relevant test would be to take the same game( ex: madden 2003) and run it on both systems. I know on the PS2 madden at times gets bogged down. During those same situations on the gamecube it doesnt. Does this mean the gamecube hardware is faster? Not really, it probably is but thats not my point. The point is the total implementation is whats important.

To all -bashers and -heads

[Anonymous Coward]

Dear Mr. [A]-basher and Mr. [B]-head,

Where A and B are,
(Mac/PC)
(Chevy/Ford)
(XBOX/PS2)
(tickle me elmo/chicken dance elmo)

or

(PC/Mac)
(Ford/Chevy)
(PS2/XBOX)
(chicken dance elmo/tickle me elmo)

Why don't you poor bastards just save up and buy one of each? Then hopefully you won't care about one more than the other and you can get on with your miserable lives.

Sincerely,
Someone who has both.

Photoshop benchmarks

[jay_in_pa]

Where does one get a copy of the scripts to run the photoshop benchmarks? I did some timings of the Watercolor filter, and was surpised by the results. Using a 800x600 scan of a photo, I got the following times (sec. speed number is bus speed): 500/100 MHz G4 : 26 seconds 466/133 MHz G4 : 26 seconds 400/100 MHz G4: 31 seconds 400/100 MHz G3: 31 seconds 1100/100 Mhz Athlon: 24 seconds 400/66 Mhz Mobile Pentium II : 31 seconds What amazed me here, is that the 400 Mhz mobile P-ii with a paltry 66 Mhz bus kept up with a G4 of the same speed. Also amazing is the G4 should no advantage of an equivalent G3. The Intel and AMD systems were both running Win2k SP2. The Macs were tested with both OS9.2 and OSX.2 (jaguar) -- results the same. Photoshop 7 was used in all tests. I'm a big fan of Macs, but also like Intel/AMD's if they don't have Windows on them, but Photoshop isn't available for Linux yet......

Re:Isn't this pointless?

[fitten]

"AltiVec" is a marketing name.

Go look up SSE and SSE2 and 3DNow! etc. and compare their goals to "AltiVec". I'm not saying which is the better implementation.

The algorithm can be copied anyway, even if there isn't direct hardware support for it, it's just vector math for which you'd use software routines instead of accessing the hardware.

Read the press release? PCs WEREN'T MENTIONED!

[Heretic2]

What the hell? What crack was the submitter smoking and the person who approved that posting. That article didn't even mention PCs or compare anything to the performance of PCs. MacBibble was ten times faster than the software THE CAMERA MANUFACTURER MADE! So what? This has NOTHING TO DO WITH PERFORMANCE RELATIVE TO THE PC!

NOTHING! This says NOTHING about performance in relation to x86. NOTHING! How could this -possibly- shed ANY light on the previous debate about performance between PCs and Macs? It's not like the previous article used any benchmarks involving software the camera people released.

*sighs* At least slashdot posts something meaningful every now and then.

Photoshop benchmarks -- repost, more readable

[jay_in_pa]

Where does one get a copy of the scripts to run the photoshop benchmarks?

I did some timings of the Watercolor filter, and was surpised by the results. Using a 800x600 scan of a photo, I got the following times (sec. speed number is bus speed):

500/100 MHz G4 : 26 seconds
466/133 MHz G4 : 26 seconds
400/100 MHz G4: 31 seconds
400/100 MHz G3: 31 seconds
1100/100 Mhz Athlon: 24 seconds
400/66 Mhz Mobile Pentium II : 31 seconds

What amazed me here, is that the 400 Mhz mobile P-ii with a paltry 66 Mhz buskept up with a G4 of the same speed. Also amazing is the G4 should no advantage of an equivalent G3. I also exptected the Athlon to shine over the Pentium a lot more than it did.

The Intel and AMD systems were both running Win2k SP2.
The Macs were tested with both OS9.2 and OSX.2 (jaguar) -- results the same.
Photoshop 7 was used in all tests. I'm a big fan of Macs, but also like Intel/AMD's if they don't have Windows on them, but Photoshop isn't available for Linux yet..

Re:Isn't this pointless?

[Alien Being]

"...American cars..."

Everyone figures that a 5 liter engine must get worse mileage than a 2 liter, but they rarely consider the fact that more torque means fewer revs.

A 400hp 5.7Liter Chevrolet Corvette gets the same 28MPG(highway) as a 140hp 1.8Liter Mazda Miata and it weighs an extra 400kg.

I beleive your talking .....

[oliverthered]

What plannet are you on?
There's the USB spec, if you follow the spec you can find out information about any device you plug in.
(like PCI)
This has nothing to do with actually using the device

A USB device will still require the device specific control, data and interrupt messages to function propaly. Thease are put into a driver, and that driver 9/10 is OS specific.

Re:I beleive your talking .....

[Jeff DeMaagd]

The public interface for a lot of standard device types is specified by the spec, including the control, data and interrupt messages. If you comply with the USB Mouse specification, then the USB Mouse driver included with the OS will work with it. For OSX, Apple already made those drivers.

That way, you don't need a specific driver by Logitech for your specific mouse, as far as the OS is concerned, the basic functionality of the mouse is the same.

ColorSync

[thinkliberty]

Macs win hands down in digital photography because of ColorSync it's easy to make sure what you are seeing on your screen is exactly what is going to print out of the printer.

Yes you can do this on a PC, but it's not as easy.

Re:Definitive Evidence

[Anonymous Coward]

Amen. /. seems to attract no one anymore who's interested in anything except their own voice when they post anyway. The clown in the original post apparently wants mommy to spoonfeed him so that he doesn't have to read anything longer than two (short) sentences.

The only surprising thing is that he didn't offer his own opinion on something he obviously knows nothing about. Personally I think they need to increase the minimum age for posts to 12.

Don't use 8.6

[Anonymous Coward]

You can't compare older Mac OSes. I use 9 at work and X at home; I wouldn't ahave bought a Mac if it hadn't been Unix. 9 freezes up plenty; mostly due to application problems. Switching from Internet Explorer to Mozilla reduced by system freezes from once every few days to almost never (on OS 9). X, however, is pretty rock-solid: 2 kernel panics in one year. Sometimes it doesn't wake up from sleep, though (this is 10.1.5).

No, it was the missing hole.

[Anonymous Coward]

The conclusion of the first article was just that, while Macs are faster at PCs at most Photoshop tasks, according to benchmarks at apple and numerous other places, RAW file conversions were the missing piece. While converters on the PC were optimized for the PC platform, reasonably good PCs were marginally faster than Macs at the task. Now that each platform has software optimized for it, the Macs are clearly faster than the PCs at this task, along with the other Photoshop filters.

I think the point is that if PCs were asked to do Quartz, they would almost certainly suck ass.

Re:No, it was the missing hole.

[Anonymous Coward]

While converters on the PC were optimized for the PC platform, reasonably good PCs were marginally faster than Macs at the task. Now that each platform has software optimized for it, the Macs are clearly faster than the PCs at this task, along with the other Photoshop filters.

The PC converters in Bibble or photoshop are NOT optimized for SSE/SSE2. Please provide us with a single reference to this.

The results were that the PC beat the hell out of the Mac with vanilla C language converters, the Mac is competitive when you compare Altivec with C, and as we've seen in so many other places (e.g. the rest of the test by Rob), the PC beats the Mac again when you are comparing SSE vs. Altivec.

BLAS

[Anonymous Coward]

Use the ATLAS BLAS on most architectures. ATLAS os optimized extremely well in x86 and Athlon CPUs, along with various workstation CPUs. It supports Altivec on PowerPC cpus. (I haven't done significant benchmarks across many architectures, but they should be available. I don't know which chips have the best performance right now.)

Re:BLAS

[UberLame]

I'll have to look into it. I generally refuse to use libraries that aren't nicely supported on linux, irix, and solaris (ideally windows and MacOS X would be supported, but really I do my primary work on linux, irix, and solaris, so if I can get my code written today, I'm willing to duplicate library code for Mac and Windows support later).

Thank you for the link.

Re:Isn't this pointless?

[fitten]

But your comparison is inherently biased since you are basically saying you are running optimized code on the G4 and non-optimized code on the P4. IIRC, the P4 doesn't require context switches to do SSE/SSE2 (not 100% sure, I know the P3s required them).

$2,420.00

[bill_mcgonigle]

That's the least you can spend on a new top-end dualie.

Based on MacBibble's specs, you should be able to convert a D1X Raw at 3008x1960 to JPG in about 2 seconds, if your storage media can even keep up.

I'm not doubting that your Sager is faster if you say it is - I'm just surprised you need it to be any faster than this.

I'll add to that

[overunderunderdone]

OK, I'm a techie and graphic designer (yes, rare).

Me too!

I generally agree though I would add a couple of points. First Apple's dominance of this particular market itself is a compelling reason for those entering the market to stick with it. If you are a designer you are going to be expected to use macs wherever you work - what kind of machine are you going to buy for yourself? If you are hiring designers you are going to be hiring people that have always used macs and might *never* have used PC's - what kind of machines are you going to buy for them? If you freelance everyone you have to work with will be sending you files from macs and taking the files you provide and using them on a mac - you can use a PC but it will add at least a little hassle. If you write software for these people which platform are you going to focus on? In the past the PC didn't even have the software required, that's changed now but still a lot of designers (especially somewhat older ones like me) still have the idea that you simply *can't* do professional design work using a PC.

Second I think focussing on stability/reliablity is a little unfair. I think windows is a lot more reliable than Mac users believe and MacOS 9 which many designers are still using until Quark gets it's act together certainly had NOTHING to brag about in that department. The real advantage is more subtle but perhaps more significant, especially considering how it apparently compensated for OS 9's UNreliablity. That is: when it worked it really did work more intuitively - This was certainly true back when the competition was DOS and then Windows 3.x and even Windows 95. After windows started to improve by borrowing heavily from the "mac way" the Mac was still more intuitively easy, partly from long familialarity by this point and partly from the continued focus on that value which shows up in Apple's focussed attention to little details that Microsoft seems to only think of as afterthoughts. I'd say OS 9 was an instance of getting all the big things wrong but getting all of the details, at least from the users perspective, right. UNIX is probably the exact opposite (which is why MacOS X is so exciting) and Windows is a (unhappy?) compromise between the two criticized from the UNIX side for getting the big things wrong (though not as wrong as the old MacOS) and criticized from the Mac side for getting the user interface details wrong (though not as wrong as UNIX). MacOS X has a real chance of getting both the fundamental things and all the little interface details right, though it's still a little immature and suffers from having had to make some comprimises. It is not quite there yet, though I think they are pulling ahead of the competition.

Re:Isn't this pointless?

[Alan Partridge]

...yeah, but the extra 400kg is what's gonna cost you the money if you ever happen to accelerate or brake the stupid thing.

Do Americans even do Physics at high school anymore?

Re:ColorSync = BS

[mattsmith]

Wrong - ICM in Windows makes it just as easy. I think that you accidentally pulled out a standard Mac argument from 1999(when ColorSync actually was better than ICM) In fact, color managment in WinXP is beter than in OSX - OSX has so many Color Management bugs that most Mac people who actually know anything about digital color matching are sticking with OS 9 or switching to Windows. Apple may have helped invent color management as we know it, but they dropped the ball over the past 2 years.

Re:Isn't this pointless?

[Alien Being]

Acceleration and braking are part of the mileage estimate, and despite the increased mass, the chevy turns in the same mileage as the mazda. It's true that the vette has a lower drag coeficient than the mazda, but it's also quite a bit wider so drag is probably about the same.

I have no idea why you would call the Corvette stupid when it's one of the world's best sports cars. C5Rs finished 1st and 4th *overall* in the 24 hour race at Daytona in 2001 against far more exotic (and apparently delicate) machinery.

Re:Isn't this pointless?

[Anonymous Coward]

Dude, I write C++. Not assembly. I don't like assembly. In this way, Altivec is better than SSE2 because I can write my code with a special C interface.

SMP

[Steveftoth]

Shouldn't the OS automatically direct the applications you are using to the 'free' processor? Aren't you being a little obsessive-compulsive about making sure that your processors are being used?
The whole point of SMP is that you as the user shouldn't have to CARE about which processor a task is running on because the machine takes care of the leg work for you.

Yes, if you give a thread processor-affinity, then sometimes it will run faster since the caches are alredy loaded for that thread. (hopefully) However, that should be the job of the OS and not the user. I mean then you end up spending time on making your computer run faster which negates the fact that you have a faster computer.

Re:SMP

[Anonvmous Coward]

"Shouldn't the OS automatically direct the applications you are using to the 'free' processor?"

Should it? Yes. Does it? No. They default to processor 0. If the app has multi-threaded support, THEN it automatically load balances the processors. (I'm talking strictly Windows here, I have no idea how *nix does this.) Not everything I do is multi-threaded, and even if it was I don't necessarily use it in that mode.

"Aren't you being a little obsessive-compulsive about making sure that your processors are being used?"

Not at all. My work flow is such that my computer can be busy for hours at a time. During those hours, I can't just sit here with my thumb up my bum. So what I do is I delegate the hard-working stuff to processor 1, and the rest of it to processor 0. This means that I can do my other stuff like answer e-mail or work on my next model or something.

It's not a matter of being 'obsessive' or wanting to be in total control, it's a matter of optimizing my work-flow. If my computer's busy, and I'm not, I get in trouble. (Yes, I know how it should work but the managers above my boss don't understand how 3d rendering works.)

Re:SMP

[PCBman!]

I believe unix oses behave similarly. At least Solaris 9 (? or was it 8) did that. Of course, we generally ran out of RAM before we ran out of CPU time.

Re:SMP

[swb]

I have never experienced the phenomenon you describe, and it goes against the grain of how I understand the scheduler works, at least in 2K.

I have seen a heavily loaded single-threaded process on a dual-CPU machine "ping-pong" between processors on a 2K machine. When you show the load of each processor in performance monitor you see a peak and valley for each CPU and exactly opposite on each CPU.

I've seen slightly asymetric loading in the task manager view, but generally it follows the same pattern. The asymmetry I've written off to task manager not having as much granularity as performance monitor.

This is exactly what I'd expect. The system is capable of running two processes or threads concurrently; the processes are scheduled round-robin onto CPUs as they are free, and the 'busy' process jumps from CPU to CPU as it becomes available and never stays on a single CPU, unless of course you force a processor affinity.

But you'd think that forcing an affinity would actually hinder process execution, since unless you told all the other processes to stay off of a given CPU, the affinitied task will get less work done as there will be fewer timeslots available to it.

Re:SMP

[Anonvmous Coward]

"I have never experienced the phenomenon you describe, and it goes against the grain of how I understand the scheduler works, at least in 2K."

I'm not going to go on record as an expert about it. It was observed behaviour. I'm not saying you're wrong or that I'm right, but that's just the way my machine appears to behave. You've got me interested in taking a more scientific stab at figuring that out.

"But you'd think that forcing an affinity would actually hinder process execution, since unless you told all the other processes to stay off of a given CPU, the affinitied task will get less work done as there will be fewer timeslots available to it."

I'm not using the dual for time-critical process, but rather for work-flow. My rendering may finish in 4 hours instead of 5, but if I'm not there when it's done then really I am the bottleneck, not my machine.

The renderings I do are in layers. Instead of rendering everything all at once (which would benefit from having both processors available like you suggest) then I have elements of a scene that render rather quickly. So what I do is I'll start one element rendering on one processor, then I move to the other processor and continue work on something else like another element or build a model or something. It may take 40 minutes as opposed to an hour to render with both processes, but that's an hour I'm still working as opposed to waiting for my machine to wrap up.

It's actually kind of neat working like that. One of the things I do is start my composite and watch it. I can tell part-way through a rendering if something is wrong, fix it, and restart it. It sure beats waiting for the whole thing to finish!

Re:SMP

[swb]

I think the most important thing would be for hardware and software vendors to re-think the whole idea of multiprocessor systems, both from a memory allocation and bus access perspective and from a process creation and CPU time allocation perspective.

It'd be almost more interesting to look at a 4 CPU system as multiple systems with varying processors depending on the needed configuration, with hardware shared or dedicated on an as-needed basis. Cluster-in-a-box or partitionable or NUMA on a motherboard or what have you.

It already exists in the mainframe/enterprise system level for IBM or Sun systems, but it'd be interesting to see a similar flexibility built into a single 4 CPU board.

Re:Isn't this pointless?

[fitten]

Something like this then from Intel? (found on google by searching for "SSE2 library"
http://cedar.intel.com/media/pdf/games/c ode_zohar. pdf

or any of the intrinsics in the Microsoft compilers (same search)
http://msdn.microsoft.com/library/default .asp?url= /library/en-us/vclang/html/vcrefwillamettefloating pointintrinsics.asp

or maybe one of the other 2,770 hits I got by doing that search.

In the worst case, you can bite the bullet and write your own SSE2 C wrappers like someone did for the AltiVec functionality?

There is no such thing as a "special" C interface.

The Trick with RC5...

[Llywelyn]

Is that is almost *exactly* the kind of task the AltiVec engine excels at. IIRC, the numbers are as follows:

P4 -- Works on one key every other clock cycle.
AMD -- Works on one key per clock cycle.
G4 -- Can work on 4 keys at once.

Or something else along those lines--regardless of the actual numbers, it demonstrates how powerful AltiVec really is.

Re:The Trick with RC5...

[PCBman!]

Actually I think it has more to do with Altivec's permute function allowing bit shifts and bit rotations all in one cycle. There was a paper about this on ars-technica.

Re:ColorSync = BS

[Anonymous Coward]

OSX has so many Color Management bugs that most Mac people who actually know anything about digital color matching are sticking with OS 9 or switching to Windows

On what basis are you making this argument? My father has worked in the graphic arts industry all his life. He got his shop converted over to OS X and won't switch back. He has absolutely no problems with colormatching, and all of the other printshops in the area he deals with still use macs (in fact their only use of pc's is to interface with hardware lacking mac drivers like digital platemakers and some presses), many switched to os x, others are in the process of switching. It doesn't appear to have affected their color-correcting capabilities. Don't make blanket statements on what the industry is or is not doing. I am not saying that you are wrong, just that it isn't the case in Western New York and that blanket statements are misleading and dangerous.

Re:Adobe Photoshop 7.0.x AltiVecCore Update plug-i

[Foamy]

I haven't used PS extensively since I installed the altivec core update, but I did notice one area where the speed increase was mind-boggling.

On my Ti550, when doing a "save for web" it used to be excruciating to wait for the optimized images to show up. Now it is almost instant. I used to dread the final steps of making a suitable web image, but now it's a breeze.

Let's not forget AppleScript and ColorSync

[Anonymous Coward]

We shouldn't forget those two apps. By using AppleScript, I can automate my workflows throughout all of my applications (Photoshop, Illustrator, etc.). That alone lets me get my work done faster than any 3Ghz P4 Dell running the latest Windows operating system. And ColorSync is crucial to any graphics designer. If the work on my screen doesn't look like the proofs I hand to my clients, and the proofs don't look like the final print I deliver, I am screwed from not only losing a client but I'm out of a job as well. No other operating system offers anything close to the capabilities of both AppleScript and ColorSync

Re:Let's not forget AppleScript and ColorSync

[Anonymous Coward]

Uh, except Windows which has VBScript and ColorSync... D'oh!

Re:Isn't this pointless?

[Steve Cowan]

quote:

The Mac program makes use of hardware only available on the mac.

/quote
---

All Mac programs make use of hardware only available on the Mac, silly.

Re:Isn't this pointless?

[noewun]

Same reason Ford was able to win the Manufacturer's Championship in 1966 and Le Mans 66-69: while the Ferraris sure were perty, the Fords, and their V8s, just kept on pulling through the night. And, yes, the whole argument is pointless.

Main problem with AltiVec

[be-fan]

AltiVec can be pretty nice, but it's got a serious limitation. The G4's memory bandwidth is a paltry 1.3 GB/sec (shared by two processors on most PowerMacs). If you consider that a single vector takes up 16 bytes, then a single AltiVec unit can theoretically chew through 16 gigabytes of data per second. Even if you take into account the G4's L3 cache (1-2MB @ 4GB/sec) the G4 has nowhere near the bandwidth of a P4, which has 4.2 GB/sec to main memory. At that point, it doesn't matter if the AltiVec unit is ten times as fast as the P4. Unless the data fits mostly in the caches, both processors will be memory bandwidth limited. This situation favors the x86 chips, because recently, memory bandwidth has been climbing very quickly. By Q3-2003, x86 chips will hit 6.4GB/sec of main memory bandwidth.

Re:Main problem with AltiVec

[Anonymous Coward]

The PPC 970 will have a 900MHZ FSB. This will even things up a bit regarding bus speed and memory

Re:Main problem with AltiVec

[tuxedobob]

Well, according to Apple, the latest PowerMacs have a "2.7 GBps throughput between main memory and the system controller". It's beyond me to postulate how this affects the stats, though.

Re:Main problem with AltiVec

[be-fan]

Yes, 2.7GB/sec between main memory and the system controller. 1.3GB/sec between the system controller and the CPU bus.

RAW format

[SPeW]

who the hell uses RAW format anyway? I've seen plenty of images files in all shapes sizes and types but I have yet to see anything that uses RAW as an image format.

Re:RAW format

[Daniel Joannidi]

Digital Photographers enjoy the RAW format over JPEG or TIFF for several reasons. A good analogy is to consider a RAW file as a digital negative, or a JPEG or TIFF as a color slide.

RAW images contain more information from the camera - they're unprocessed, like a digital negative. JPEG's will have much of the same information, and with a low compression ratio will often have similar 'quality'. When you bring these into Photoshop and try to modify or play with the pictures, a RAW file will give you more information to fiddle with.

Rob Galbraith explains this in greater detail
I've included some relevant quotes below:

Because RAW photos are mostly unprocessed in the camera, the white balance, hue, contrast, sharpening and exposure settings can be overridden in software after the fact. All but exposure can be overridden completely; that is, the resulting processed photo will look exactly as if the photo was shot on the correct settings in the camera in the first place. Most RAW file processing apps, with most RAW file formats, allow for great underexposure recovery: shoot the picture at ISO 200, underexpose by 2 stops, use the magical software exposure compensation control to brighten the picture 2 stops, and in most respects the processed photo will resemble an ISO 800 photo straight out of the camera.

(snip)

So, another measure of image quality is the ability to fix white balance, exposure and other errors after the fact, in a manner that is vastly more effective, and quicker, than what could be accomplished in Photoshop with a JPEG or standard TIFF. RAW allows news photographers to continue to shoot colour negative in effect, with the same sort of latitude for various types of technical errors, instead of having to switch to the more unforgiving colour slide, which is what JPEG or standard TIFF conceptually. This analogy isn't perfect, because unlike colour negative film, which doesn't offer the same overall quality of color slide, RAW photos inherently offer better quality than JPEG or standard TIFF.

Re:RAW format

[SPeW]

OK sounds good, but how many digital cameras let you get that RAW file most that I know just give you the JPG. At least for commercial or home use, I'm sure that there's a few professional models that let you select the format it captures in.

Re:RAW format

[Anonymous Coward]

Most actually (dig through the menus or RTFM).

The bigger problem is image size. Pun intended. RAW takes up a bunch more space than JPG.

mac drivers and applications

[toothfish]

the lesson from this seems to be that, given the mac's proportionally smaller user base, mac users seem to often get the short end of the stick when it comes to cross-platform drivers &c. the makers of vuescan and macbibble can probably attest to this. bibble, as i recall, was originally inspired by nikon's hideously buggy/crash-prone driver software for their D1, which set a few bars in terms of digital camera hardware but was well-nigh unuseable by mac users because of the software (which you had to buy separately!). we're seeing the same thing these days with quark, albeit in stronger language.

granted, this is pretty much the way it's going to be as long as mac users make up 5-10% (depending on who you ask) of the desktop computer using population, but it's still a drag.

Tools are tools

[Anonymous Coward]

Computer choice should never be about speed comparisons. Its about the righ tool for the right job. I run ProTools LE at home. Newer Athlons smoke any Mac (and P4s) for Host-based processing in ProTools. So, in this case, an Athlon is the right tool.
When I upgrade to a ProTools TDM system I'll build it around a Mac because the 3rd party support is much better for this Native system. In this case the Mac is the right tool.

Additionaly the only people I know where speed makes THAT much of a difference are gamers. Everybody else could well use a Mac.

AC

Re:New phrase coined:

[PetWolverine]

That said, maybe the word "death" is too negative. It's more of a Beachball of Thought, or a Beachball of One of These Words. (My favorite: Beachball of Rumination.)

Then, when it appears and won't go away, because a program has crashed, it's just a Beachball of Eternal Rumination.

Now, when you get the white-on-black text scrolling down the screen and the mouse cursor disappears, that's the time to start throwing around the word "death".

Macs are slow and expensive for cpu intense ...

[omegadva]

processes but fast in terms of the actual work you get done.

There is no need to take an either or attitude to PC vs Mac. They play well together...

I do a certain amount of video editting. Doing the actual editting is faster on the Mac with Final Cut Pro vs Premier / Pinnacle / Ulead on the PC ( I've tried them all in an effort to stay with only one platform.)

Now doing the actual rendering on the Mac is slow....... Check out
mac vs HT PC
The above article far understates the true advantage of Intel hardware. To truly match the
Apple architecture with Intel hardware. Go to
Dell and configure a dual 2.8 Ghz Xeon Workstation with Hyperthreading on each chip. Now you have 4 processors working for you , two real and two virtual per CPU. For $3200 you get two 2.8 Ghz Xeons with 1 gig of RAM, 2 80 gig harddrives (separating the system disk from your video disk is important) and a 4x dvd+rw drive.

Still think the G4 dual processor Mac is great? Why not use the industry standard to measure your chip of choice. In the supercomputing world how fast your machine runs is more than just bragging rights, it's job security. For that reason the SPEC benchmarks were created to get standardized validated results on any hardware. Mac OS X has a SPEC suite see
Mac SPEC

Now that you've looked at that go to SPEC and look at the CPU benchmarks. Note the scaling factor for the Xeon 2.8 Ghz with two processors
SPEC CPU
Excerpt of CPU INT multiprocessor
Chip Result
2.8 GHz Xeon 10.2
2 CPU 2.8 GHz Xeon 18.0

The SPEC is designed to show good scaling with parallelism (multiple CPU) and here shows a 1.8 scaling factor.

So your Dell machine with HT will have a greater than 2 scaling factor for highly parallel processes.

NB: I only have a 2.4 Ghz P4 as my rendering machine and it's still faster than the Powermac by enough to make me stick with the mixed network approach.

So work on the Mac -> DV over gigabit ethernet to a multiprocessor Intel dual processor machine that renders AND burns the DVD, VCD, SVCD faster.

We all know Macs are better at what matters during the creative process. Let the Intel hardware bear the drudgery ;-)

Re:Macs are slow and expensive for cpu intense ...

[pressman]

Or better yet, do all of your editing in MiniDV in Final cut in OfflineRT mode, export an EDL, take the timecode data into the post house of your choice and have them recomposite the high rez footage inside one of their HD telecine suites. Some of the Sony HD telecine suites out there will put just about any home based Intel renderer to shame. Even some of the SGI Flame and Inferno systems will put most systems to shame in terms of render time.

Granted, you break away from the DIY model, but who can afford to have an HD telecine suite in their home? I mean, come on, most /.'ers don't want to pay for anything, let alone a $2 million proprietary hardware/software suite from Sony!

(ducks and covers!)

Re:Isn't this pointless?

[Alan Partridge]

one of the world's best sports cars?

I have NEVER seen one on the road, yet I see about 50 MX-5s, 30 Porsche 911s, 20 BMW M3s and sundry Ferraris, MGs, Maseratis, Aston Martins, Lotus', a Honda NSX, a shitload of Mitsubishi Lancer Evo IV, V, VI and VIIs and a boatload of Subara WRXs when I drive my MAZDA RX-7 to work and back every day.

Re:Adobe Photoshop 7.0.x AltiVecCore Update plug-i

[Anonymous Coward]

It does not contain any speedup, it is to address a potential crash in the existing AltiVecCore plug-in. That's it.

Re:Isn't this pointless?

[noewun]

I see Corvettes all the time. Maybe it's a geographic thing.

Re:Isn't this pointless?

[Anonymous Coward]

This is a good reply. I have access to several optimizing compilers for intel, including the free-as-beer linux one. I had access to the trial version of the intel plugin for vc++ but it expired. On one level I can turn on all sorts of compiler switches, and I can get a 50% performance boost. This all comes from compiler directives, working with C/C++. To get a really magical performance boosts I need to rewrite my code to do the expensive parts in SSE1 or SSE2, no compiler is going to do this for me.

Enter apple - they have a version of gcc with "c" commands that directly compile into altivec vectorized instructions. In a day and a half I rewrote 20 lines of C code into a big section with #ifdef __ALTIVEC__ and, boom, I had nearly 4x faster code than I had without the statement. Altivec statements map directly to assembly commands, so I was in effect writing assembly code, but at a very high level.

Don't get me wrong, Intel is way faster for more than half of the types of problems I work on. SSE2 also works on doubles. Altivec doesn't. I have three intel computers and one powerbook. Monte Carlo is the only thing so far that the mac is faster at. Well, that and ripping mp3's as compared to the linux box. I will check out your links but if they involve any magic binaries or special compilers, I doubt I will ever get around to it.

Re:Isn't this pointless?

[Anonymous Coward]

I should point out that I normally can write many more than 20 lines of code in a day and a half.

Re:New phrase coined:

[jo_ham]

10.2 doesn't panic like that any more, that was just 10.1.x.

Fortunately, I haven't seen any panics in 10.2 yet, but I have seen pictures of the screen that shows up if it happens (taken with a digital camera - for some unknown reason you can't take a screen shot during a kernel panic...) and it says something like "Your computer has panicked. You will need to reboot it. You will lose all your unsaved work" or words to that effect.

Oh, and I'm cracking up at Beachball of Rumination, that has forever entered my vocabulary now. Formerly I was calling it the spinning beachball of death.

Re:Isn't this pointless?

[Alan Partridge]

perhaps you missed ny point - why do only Americans buy Corvettes?

Is it because-

a - they're shit cars
b - Americans wouldn'tknow a sports car if they ran over one in a Ford F-150

YES IT IS!

[tshak]

It's not about speed.

Yes it is!

Sure, for those who browse the web and edit small image files, no, it's not about speed. But my hobby is audio (some video) production. My Athlon1.2Ghz barely keeps up with some complex software synthesis packages (extremely complex Reason songs). I want to eventually move all of my A/V stuff to a Mac, and use a MOTU 828 firewire audio interface. Unfortunately, I'd need a Dual1ghz G4 - this costs me almost twice as much $$$ as the Athlon XP 2600+ which I'm about to build which in some ways (because of the 333FSB) may run a bit faster. So, I could either get a ~800mhz G4 which would not be fast enough for me, or I can get an Athlon XP2600+. True, the OS (WinXP) isn't quite as nice, and the hardware isn't quite as slick, but if my audio is clipping because there's too much CPU usage, the Mac solution is useless to me.

Re:YES IT IS!

[hcdejong]

The point the original poster was trying to make is that CPU speed is the bottleneck in only a minority of current uses for computers.

And if CPU usage causes your audio to clip, then the software and/or the audio interface sucks. If Digidesign could run 48 channels of AD conversion on a Mac IIci without clipping or dropping samples, then surely your current system should be able to do this.

Re:YES IT IS!

[tshak]

The point the original poster was trying to make is that CPU speed is the bottleneck in only a minority of current uses for computers.

I understand, but with MP3 and DVD encoding, Video Editing, and Gaming I wouldn't say that the minority of users don't need a fast CPU.

And if CPU usage causes your audio to clip, then the software and/or the audio interface sucks.

No, you are talking about simple hard disk audio recording. Of course I can do that on an iMac G3. I'm talking about software synthesis (physical modeling, software sampling, DSP, etc.) which can eat up a LOT of CPU when using multiple instruments and effects simultaneously and in realtime from a MIDI controller. For the specific app I'm talking about (which is nicely optimised for OS X and dual proc operation), check out Reason.

Re:Isn't this pointless?

[Anonymous Coward]

But your comparison is inherently biased since you are basically saying you are running optimized code on the G4 and non-optimized code on the P4. IIRC, the P4 doesn't require context switches to do SSE/SSE2 (not 100% sure, I know the P3s required them).

No x86 CPU has ever required a context switch to use SSE. The Pentium3 required a state switch for the FPU, which in some cases could cost 30-40 clock cycles (in comparison to a context switch between processes which would cost 1000's of cycles). Start by downloading the lastest version of the Intel compiler and ask it to optimize for SSE2 with "-O3 -tpp7 -xW". You might be surprised what the compiler can do automatically for you.

Re:Isn't this pointless?

[Anonymous Coward]

using altivec extensions is much easier than SSE2 extensions;

OK, please tell me why Altivec is so much easier (apart from the fact that it cannot do double precision):

Altivec:
Compile with -faltivec:

c = vec_madd(a,b);

SSE:
Include "xmmintrin.h":

c = _mm_add_ps(a,b);

There is no context switching involved when using SSE (you do know what context switching is, right?), but on the Pentium3 you have to empty the registers at a cost of about 30 cycles. In practice this is pointless, since nobody uses both the FPU and VPU simultaneously on the G4; it is dead-slow to transfer data between the two units and since it can only retire 2-3 instructions per clock you would starve the VPU if you tried (if you haven't tried it, run your code through sim_g4 ; it is an interesting experience!)

Now, *design-wise* you might claim that Altivec is nice, and I agree with you there. The only problem is that those 128-bit wide data paths is one of the most important reasons why Motorola haven't been able to scale the frequency of the G4. Say what you want about intel, but they aren't stupid. The main reason they stuck to a 64-bit data path even for SSE is probably that the realized that the final performance would be better due to the higher clock.

Re:New phrase coined:

[PetWolverine]

Actually, I've seen the scrolling text in 10.2, or I wouldn't have mentioned it. Only once, though; usually it does what you were probably talking about: a little box comes up saying, "You need to restart your computer now." in about seventeen zillion different languages, the rest of the screen dims, and...the mouse cursor still disappears.

More annoying to me is that this little box appears on my secondary monitor...I'm not sure why, but this is unfortunate as that monitor is an old CRT very much lacking blue phosphors, which makes it rather dark and gives it a yellowish cast. I wish I could see the kernel panic in its full glory!

This is also the screen on which the scrolling text appeared that one time, so maybe it's just because it's the leftmost screen.

Oh, BTW, I'm running OS X 10.2.3 Server...maybe the server edition still has the weird text-scrolling panics occasionally...more likely it's a bug.

As far as screenshots...when the kernel panics, nothing functions any more, not even Grab or whatever that program is called. I do have a screenshot of the BSOD, though--cropped from a virtual PC screenshot. Every time I use Windows I get it, so now it's my desktop, to remind me never to use bad OSes.

Re:Read the press release? PCs WEREN'T MENTIONED!

[tuxedobob]

Now click on one of the other links in the story and find out.

Re:Isn't this pointless?

[Alien Being]

No, I understood your point just fine. Your a non-American with an inferiority complex.

BitchBlaster 2023

[Anonymous Coward]

Ahh, the classic games...

DPReview Announcement

[macmurph]

DPReview covers the MacBibble 3 announcement here:

http://www.dpreview.com/news/0301/03012901macbibbl ev3.asp

Re:New phrase coined:

[jo_ham]

heh, my "can't take a screenshot" was a bit of a quip. Much like asking a dead man to tell you what killed him and how he's feeling.

Rob Galbraith follows up on his article...

[caleugene]

http://www.robgalbraith.com/cgi-bin/ultimatebb.cgi ?ubb=get_topic&f=29&t=000009 Indeed. Looks like he jumped the gun with his original article. With MacBibble, the dual 1.25 GHz Mac handily defeated the single 3.06 MHz PC for once...

Re: clickable link

[caleugene]

Rob Galbraith now says Macs faster...

Sorry, got to used to UBB automagically making links clickable.

Forget this

[Anonymous Coward]

Just take a $3000 Apple with dual G4 and benchmark it against a $3000 dual Xeon workstation. Everytime people make benchmarks the Dell or whatnot opposite the newest mac is usualy half it's price... What does this prove. Nothing.

I got a dual p3 1ghz and it's over 2 times faster than my G4 450. Duh... I would hope so it has twice the number of CPU.

Pointless.

Re:Isn't this pointless?

[Anonymous Coward]

""AltiVec" is a marketing name."

No, "Velocity Engine" is the marketing name Apple use. "AltiVec" is the correct term, as used by Motorola and IBM.

Nude Photos of Morgan Webb

[Anonymous Coward]

But Overall isn't a dual 1.42 Ghz G4 processor Mac faster than a 3Ghz P4 when the P4 system is choked out with the latest virus/worm/MS security flaw of the week?

Stick that in your deep pipeline and smoke it.

Re:Isn't this pointless?

[Anonymous Coward]

you are = you're

dimwit