Slashdot Mirror

I once saw this (the legacy hooks in Quartz) documented in detail on a GnuStep mailing list, but I can't google it and their archive does not seem to be searchable.

My Apple tech reps don't want to discuss it, or anything else interesting for that matter. I did find a mention of it on Ars Technica and Planet PDF.

The specs of the system and the money involved have very, very little to do with the overall quality of the games. The Sega Genesis had jack for money and barely any developers against the NES behemoth. They didn't need money, they didn't need to "entice developers," they got a few good ones together that would work for cheap, made their own games for their own system, and gave Nintendo one hell of a run for their money. One could even say that the infusion of cash has been a bad thing for the industry, large markets have turned something that was very much an art form for decent profit to a big media market. When anything that looks somewhat pretty or is halfway entertaining makes a metric assload of money, there's much less incentive than there was when only the really great games got any sizable amount of cash. There's just no motivation anymore, which is why we can always predict that the great games are going to come from people like Yuki Naka, Yu Suzuki, Shigeru Miyamoto, their teams, and the projects they supervise. The names are so important because these are the people that still treat it like an art (except Yu Suzuki not so much anymore, but he hasn't had any public support since Shenmue flopped).

Also, it's pretty silly to call a console with UMA technologically superior to the others. Just because they throw fancy/powerful PC hardware in a box doesn't mean it's a superior gaming platform. The biggest point of a console is architecture, not raw power, not how impressive the hardware is, but how incredibly well everything works together. One of the best things Sony did with the PS2 was the emotion engine, while it had some serious oversights (lack of hardware antialiasing), it was a better attempt at a SYSTEM than MS's PC in a slightly smaller box. I don't know as much about the customization of flipper and the rest of the gamecube, but from what I know of the N64 I think I can assume that Nintendo knows what they're doing hardware-wise (aside from that unfortunate incident with the N64 cartridges). I know that someone's going to mention all the wonderful benchmarks that everyone and their mother with a dev kit has put out. Be sure to read what they're testing - 9 times out of 10, the benchmarks are just polygons without effects. Benchmarks on consoles are useless.

Actually, he's right... at least in the original sense of the term. Here's an Ars Technica page on the subject. Basically, back when the concept of RAID was developed, it was the alternative to a SLED, or Single Large Expensive Disk. Thus, a RAID was, most assuredly, a Redundant Array of Inexpensive Disks. It's been gradually bastardized into "independent."

The Klatz-2 implemented this technology... Here's a story on it and a link to their wiring app...

You are correct that Crusoe does the translation in software. Aside from that, it's the same idea, and it makes the ISA largely irrelevant for the same reasons.

Of an Ars Technica article posted on the 26th of October, that you can still read here

the length of pipeline does not a processor make

If you are suggesting that the P4 has a better pipeline design than the G4 or the (granted 10 month out) PowerPC 970. You will be disappointed by this article: The Pentium 4 and the G4e: an Architectural Comparison

Heres an education:

Shorter data pipeline
The performance advantage of the PowerPC G4 starts with its data pipeline. The term "processor pipeline" refers to the number of processing steps, or stages, it takes to accomplish a task. The fewer the steps, the shorter -- and more efficient -- the pipeline. Thanks to its efficient 7-stage design (versus 20 stages for the Pentium 4 processor) the G4 processor can accomplish a task with 13 fewer steps than the PC. You do the math.

All advanced processors try to guess what they will need to do next in order to increase performance. This is known as "speculative operation." Of course the processor doesn't always guess correctly, and when it's wrong it must often clear out the pipeline and start over. This results in bubbles -- or periods of time where no data is available for processing -- that leave the processor idle while it waits for new data. Because the G4 pipeline is short, the processor recovers from bubbles more quickly, resulting in higher processor utilization. With fewer processing steps, faster recovery and higher processor utilization, processor output is maximized.

Another aspect of speculative operation worth noting is that it is possible to create (for testing purposes) a contrived set of instructions that can make the processor guess correctly much more often than it would under real-world conditions. Thus a "benchmark" with no relation to actual performance can be crafted to cleverly avoid the bubble problem and thus indicate unrealistically high performance. This underscores the importance of using real applications to provide valid performance comparisons.

-Apple Computer

Regardless of how slow Mac OS X does or does not 'feel', applications do have a tendency to respond and startup slowly. Some figures (measured right after logging in on a iBook 700Mhz/384MB/20GB using Mac OS X 10.1.5):

mozilla 1.2 beta startup:
first run: 13s
second run: 6s

chimera (the 'fast' mozilla) startup:
first run: 8s
second: 2s

terminal application (the console):
first run: 11s
second: 2s

MS Word: 6s, XDarwin: 24s, etc...

Even compared to far slower PC's running Windows or Linux, I find those figures nearly unacceptable. Bouncing icons are neat to look at, but they do get boring after a while...

Aside from these figures, there _is_ the subjective OS X 'feel'. I noticed a lot of posts talking about how 'nice' and 'fine' OS X runs (on G4 CPU's with lots of RAM, of course). That's just how OS X feels like: it's responsive 'enough'. When you click something: it almost instantly responds. Almost.
Luckily, while all the eye candy is heavy on CPU load, Apple made sure that Mac OS X gives you feedback enough to make sure that you know it's doing _something_ (the dreaded pinwheel of death excluded).
Mac OS X definitely misses the snappiness of Windows or Linux though... but I guess a lot of users percieve this as a 'stable and solid' feel.

Oh and, Jaguar reportedly cuts off approximately 1-3 seconds of the mentioned startup times. But shelling out more than 100 bucks to make my apps start a sec faster? I don't know about that...

By the way, a very interesting read about the performance in general of the Mac OS X versions can be found here.

you might try the Ars Technica team if you want to win the prize ;)

It's possible to do graphics in OS X without using Quartz.

I'd be very surprised if that's the case! Quartz interfaces with the I/O Kit, which is what talks to the hardware. It provides the drawing context for everything that appears on the display.

Aqua is the entire user interface (the windowing system is a subset of the user interface).

Sorry to be difficult, but I don't think that's entirely accurate. If you count the Core Graphic Services as part of Quartz (Apple does), then OpenGL, Quicktime and [Carbon] Quickdraw all depend on Quartz. Ars Technica has a good overview of the relationships.

This is only relevant because you could replace Aqua with a different interface using Quartz. Without Quartz, there is no Aqua. The reverse isn't true.

Hyperthreading works well for certain types of software, and awful for others.

Here's an article from Ars Technica on HT/SMT.

Well, Apple seems to be weaning away from Motorola and going to IBM for the IBM PowerPC 970 (A nice tech-spec here). Yet it's still a PowerPC architecture.

Something tells me there are more actions at play here than simply changing processor manufactures in this x86 release. What exactly, I don't know. But still, there's something more here than meets the eye.

If I cancel my account, how am I supposed to paypal Cezar teh moniez?!?!

Because it's a reference to Ars Technica and its Imperator, Ken "Caesar" Fisher.

Finder is the most efficient (IMHO) file management system and perfected in Mac OS X.

Ars Technica disagrees with you. Even in Jaguar, they still have problems with the OS X finder. Maybe you feel that Finder was perfected in OS X, but judging by Ars Technica's many reviews (and their statements that much of the Mac community agrees with them about the Finder issues), I have to assume that you're in a very small minority.

Finder is the most efficient (IMHO) file management system and perfected in Mac OS X.

Ars Technica disagrees with you. Even in Jaguar, they still have problems with the OS X finder. Maybe you feel that Finder was perfected in OS X, but judging by Ars Technica's many reviews (and their statements that much of the Mac community agrees with them about the Finder issues), I have to assume that you're in a very small minority.

Finder is the most efficient (IMHO) file management system and perfected in Mac OS X.

Ars Technica disagrees with you. Even in Jaguar, they still have problems with the OS X finder. Maybe you feel that Finder was perfected in OS X, but judging by Ars Technica's many reviews (and their statements that much of the Mac community agrees with them about the Finder issues), I have to assume that you're in a very small minority.

Finder is the most efficient (IMHO) file management system and perfected in Mac OS X.

Ars Technica disagrees with you. Even in Jaguar, they still have problems with the OS X finder. Maybe you feel that Finder was perfected in OS X, but judging by Ars Technica's many reviews (and their statements that much of the Mac community agrees with them about the Finder issues), I have to assume that you're in a very small minority.

I found some G4 specmarks here.

What metadata?

So with a journaling file system, file metadata will be stored at the filesystem level, not as a component of the files themselves, right? If my understanding is correct, this kinda makes John Siracusa's whining rant about Apple's metadata in OS X being all wrong kind of...pointless. And dumb.

Could it be that maybe, just maybe, Apple had an eye to the future when developing it's file metadata structure? That it's entirely possible that Apple has had this in development for some time and there is a reason they changed their metadata structure? Siracusa seems convinced that it's inconceivable that Macs would ever support a filesystem like the proposed Elvis system and therefore everything they've done to OS X (metadata-wise) is wrong.

I love this new development simply for the reason that it spits in Siracusa's face. He's generally a pretty knowledgable guy, but his metadata article has rubbed me the wrong way for a while and it's good to see that, while accusing Apple of not being forward thinking, he was the one not using the old gray matter.

Actually... to the best of my knowledge that uninstall method has issues.

I think those issues pale in comparison to the OS X meta-data and directory separator travesties.

Now here's a case of when life hands you lemons you make lemonade. The parent was talking about simultaneous execution, i.e. how many instructions per cycle can come out of the end of the pipeline. You're twisting it around to take that number and multiply it by the horribly long pipeline.

Let's go back to basics, every time the processor makes a mistake in guessing what's going to happen next, the pipeline has to be cleared. Every modern CPU faces this problem so you want short pipelines so your penalty is low. Intel has vastly longer pipelines and thus they pay a higher price every time predictive branching screws up.

So having a large number of instructions being simultaneously worked on is a *bad* thing unless they are also being pumped out and executed in large numbers as well. AFAIK, in the P4 they aren't.

According to Ars Technica the P4 in the real world gets 2.5 instructions per cycle done. With the new G5 getting 8 done per cycle with half the pipeline depth, performance should once again favor the Mac side of the PC wars.

And what about conditional branches nearby? You don't know until the instruction commits what the register names will be. Imagine code which simply conditionally branches around a remap instruction. How do you handle that sanely?

I personally think the remap idea is insane. You're essentially adding register-file mode bits, and mode bits are just ugly in too many ways to describe. Just add more architectural registers already! The Pentium 4 has 128 rename registers anyway, so it seems like adding more 'architectural registers' is more an opcode formality than anything else.

And you don't understand how modern processors really work. First, they have several levels of cache memory exactly because you don't want to go out to main memory too often. Second, they have many more registers than the assembly instructions can see: register renaming, speculative eeecution and all those tricks reorganize instructions so that the CPU core doesn't really have to move data back and forth between memory and registers so often.

IANACPUD (I'm not a CPU designer) so I'm not going to try to descripe this stuff further, but articles abound. Here are some: Into the K7, Part One and Into the K7, Part Two

And you don't understand how modern processors really work. First, they have several levels of cache memory exactly because you don't want to go out to main memory too often. Second, they have many more registers than the assembly instructions can see: register renaming, speculative eeecution and all those tricks reorganize instructions so that the CPU core doesn't really have to move data back and forth between memory and registers so often.

IANACPUD (I'm not a CPU designer) so I'm not going to try to descripe this stuff further, but articles abound. Here are some: Into the K7, Part One and Into the K7, Part Two

Sometimes I can't figure out WHY people refuse to research themselves.

This might be a good question, "Where did all the hand-held scanners go", but, you can learn about the whole history in a few hours of seraching on the net.

In a remarkable feat of journalistic sleight-of-hand, thousands of column inches in many "reputable" on-line publications have talked at length about .NET whilst remaining largely ignorant of its nature, purpose, and implementation. Ask what .NET is, and you'll receive a wide range of answers, few of them accurate, all of them conflicting. Confusion amongst the press is rampant.

... or don't we see chipset manufacturers avoiding the hard problems completely? I realize that cost is an issue, but for the most part, we're talking about high-performance workstation and server boards, which cost $500+ or more.

The biggest issues these days are:

• Data Starved Processors - and this is all about latency (and, to a lesser extent, bandwidth) to main memory. I don't care if there is DDR400 memory support, what I want to kow is why isn't there a L3 cache? I mean, even the high-end Xeons these days have a max of 1MB or so on-die L2. Sure that's great, but do you know how many common datasets blow right through that? It's often dozens (if not hundreds) of cycles to access main RAM. The alpha architecture did L3 on the motherboard way back in 1994 or so. Why don't these modern server chipsets support 16MB or so of SRAM for L3 cache? Hell, they should probably support 64MB or so.
• Improved hyperthreading support - go check out the Ars Technica article on this. Hyperthreading can potentially really help performance, but it's being held back by (among other things) problems with cache coherency and loading. While much of this is on the CPU (and thus, a chipset can't help), there are a bunch of stuff that could be moved into the chipset for help.
• Useful shit in the Chipset - ATA/133 isn't that useful (vs ATA/100). Firewire is OK, as is USB 2.0, but what I want to know is where are nice stuff like block data copy between video and RAM (like the SGI chipsets for the Indy/O2 had) for high-performance video processing? AGP is a joke for this (as anyone doing video processing will tell you). These chipsets are aimed at workstations, after all.
• Standard interfaces for custom silicon - no, I'm not talking PCI-X or crap like that. There should be a standard interface directly to the chipset for people who want to do custom silicon ASICs and have them have direct access to the high-bandwidth internals of the chipset. I mean, even in the low end, why should a FCAL controller chip have to pass the PCI bus? Or a hard-core encryption coprocessor? Or a hardware routing ASIC? All need several GB of bandwidth directly to memory (or each other), and I can't see any reason not to have them surface mounted next to the north bridge with a dedicated interface.

Unfortunately, there seems to be little innovation going on in chipsets these days. The high end looks very, very, very depressingly identical to the cheap consumer crap. WTF folks?

-Erik

There's a slightly more technical writeup of the API dilemma in John Siracusa's Ars Technica review of Jaguar.

Here's the beginning of the article.

There's a slightly more technical writeup of the API dilemma in John Siracusa's Ars Technica review of Jaguar.

Here's the beginning of the article.

Ok, maybe I need a detailed explanation.

I'll be over here reading while you write:

P4 and G4

Macintoshian Achaia

Thanks...

Read more about it on a site that /. seem to respect, Ars Technica. Good article by John Siracusa: Developers, Developers, Developers!

John Siracusa's Mac OS 10.2 review on Ars Technica also covered this. See API Wars

Looks like Apple are trying to stop the tweaks that make Mac OS X usabale. Which seems daft to me.

I think this Hot Rod configuration gives way better value for money. Pretty damn good system for $1600.

And people get excited about War Flying... Heheh you have no idea what you can do with the right equipment. The air around is us filled with information, all you have to do is listen.

::sigh:: This is what I'm talking about. Apple's design sucks. Yes, it allows for fancy translucent windows, but it puts a significant burden on 3D hardware that is already memory-bandwidth limited to begin with.

Second, because PDF is used as an internal representation, OS X can't easily use OpenGL for the "real work" case of accelerating all Quartz 2D drawing commands rather than just window-compositing.

Other designs like Longhorn, EVAS, and Berlin, do not have this limitation.

The G4 Cube Apple made was a good start in that direction, as was the iMac.

With a central "convection column" we could put the processor low in the box (it would need a stand like the G4 Cube to allow airflow underneath) and position components around the column, we might be able to do it.

Of course, if you just want to leave out fans, and don't want to explore liquid cooling, you could use Peltier effect (Ars Technica has some details) coolers with heat sinks and the "convection column" or a heat distribution "tree" that spread heat out along sinks until it could be expelled along the case sides...

It's possible, it would just take more effort than many are interested in.

Of course, you could always pipe Central Air into your case...

Check out this article on bios settings.

This is where our hero Microsoft comes in. They wil make slower OS that gobbles up more CPU and RAM so that these fast Intel and AMD processors will sell.

a) can you specify in how far they're slower? GUI, network performance, Application launching?

b) If you're referring to the GUI (likely!), well, that's because the Quartz GUI is actually a whole new approach that's simply more demanding than Win2k's old-generation GUI! They did, however accelerate it alot in 10.2's OGL-accelerated Quartz Extreme, you should really check that out! I can run the iTunes Vizualizer and 2 transparent browserwindows and a transparent terminal on top of it with no fps-drops whatsoever.. on a 700MHz G3 iBook with a measly 16MB Radeon M6! Windows still need to be rendered by the CPU initially, but all the compositing is offloaded onto the 3D-Hardware! That's something that's due for Longhorn somewhen after 2005!

c) If you really wanna compare, be fair and compare to WinXP, which has GDI+ (non-accelerated yet!), which is a classic blatant M$-steal (again!) of Apple's Quartz concept! And XP's GUI sure is just as slow as Aqua was before 10.2, just be fair and don't compare G4-400s with 1.8GHz Athlons or 2.8GHz P4s!

I agree... the Metadata usage in OS X is a step backwards. But it's actually that they're coming from the Mac method (type/creator codes) to the Windows method (with the unix layer using the unix method, obviously). They really should have added "chtype" and "chcreator" utilities ao the BSD layer - and submitted mods to "ls" to optionally display that info from the command line. =(

Try this article at ArsTechnica for more insight: Metadata, The Mac, and You.

For those of you that missed the link at the beginning of the article, take a look at Apple's full Aqua HI Guidelines (or in PDF format). It has *tons* of specific examples and screenshots useful for some of the theory and design behind the current GUI.

I have to agree with the earlier post that OS X is somewhat of a step backward in usability overall. Although I do appreciate some of the innovations (sheets,...) I still find the standard OS 8/9 "platinum" interface to be easier to understand. (It's an interesting comparasion.)

I don't know -- once Apple gets its butt in gear and gives me a SPACIAL FINDER and uses METADATA PROPERLY I might feel different. :)

For those of you that missed the link at the beginning of the article, take a look at Apple's full Aqua HI Guidelines (or in PDF format). It has *tons* of specific examples and screenshots useful for some of the theory and design behind the current GUI.

I have to agree with the earlier post that OS X is somewhat of a step backward in usability overall. Although I do appreciate some of the innovations (sheets,...) I still find the standard OS 8/9 "platinum" interface to be easier to understand. (It's an interesting comparasion.)

I don't know -- once Apple gets its butt in gear and gives me a SPACIAL FINDER and uses METADATA PROPERLY I might feel different. :)

This page is the best, it mentions the coolest screensaver I've ever seen: Marine Aquarium. And it describes other fun things you can do leverage the power of Quartz Extreme. I'd love to have a QE compatible video card to enable this screensaver as my desktop using that nice little comannd line: /System/Library/Frameworks/ScreenSaver.framework/R esources/ScreenSaverEngine.app/Contents/MacOS/Scre enSaverEngine -background

Looks to me like leakage from your AP might be even easier to get a look at from above.

http://arstechnica.com/wankerdesk/3q02/warflying-1 .html

My next assumption is that anyone silly enough to be using wireless without decent encryption at the network layer will soon be getting sniffed at the satellite layer! Cheers.

Most of the people on this discussion seem to think that hybrid cars only the transition to fuel cell vehicles. Although, I probably tend to agree, I question when exactly fuel cell vehicles will actually be available.

I actually bought a Honda Civic Hybrid last week. I've gotten just under 1000 miles on the car now and the car is only on its second tank of gas. The readout shows 43.1mpg at the moment.

The car has surprisingly good power both off of the line and at top speed (when you need that extra ooomph to cut off the guy in the fast lane to get around the slow guy in front of you).

ArsTechnica has a good article about it. Especially, check out the CVT transmission. You don't even feel the car shift.

I don't work for Honda or anything like that (as I realize this sounds like and ad), but I love it so far, it gets awesome mileage, and there's even a $2000 tax break for owning one.

Even if it is a "stop gap" until fuel cells show up in mass quantities and reasonable prices, I'm very happy with my hybrid for the time being.

You did spell oppressive as "opressive" in CID;
http://ask.slashdot.org/comments.pl?sid=38792&cid= 4161583; but that will not derail your argument or distract me from reasonable discussion. That*s two "p's," not one.

I personally use my real account and meta-moderate with it. I make every effort to make anything modded down and mark these as "unfair" without looking at the moderated post. It's almost similar to how I vote in "real life." Since categorically all politicians are corrupted to some degree, I screw the incumbent out of office if I know no great love for him. I may be doing the public a disservice in both cases, but 9 times out of 10 I'm not. Moderators here classically are very aggressive to dole out points in a negative fashion. Humans by nature are like this. They would rather trample on others to "get up." I meta-moderate all positive moderations as "fair." This is because there is not enough love here. If "-1" was truly reserved for just crap (disgusting ascii art, offtopic, crap flooding), I would not feel this way. But I feel compelled to read comments, if at all, at -1, because most of the wittiest and acerbic humor and insights come from the bowels of the comments." +5", which I can obtain with almost certainty on my real account [but have now chosen not to be a boot-licking serf anymore and start expressing disgust], is awarded to extremely crafty humorous fast responses, which are rare. The rest of the "+5" high mods are generally doled out to linkers, story re-posters, or Katz like ranters who are pseudo intelligent, horribly deficient in geopolitical knowledge or world politics, ultra left wing idiots, SUV hating dip-shits (eg; those who don*t blame fossil fuels in general but just one vehicle that burns fossil fuels, but they would charge an electric car from a coal factory which is less efficient that an SUV) and other sorts of geek-socialist types. Lots of decent commentary in 3-4 range. Because most people can post at +2, there is a lot of crap in the 1-2 range. But "+5" to me always seems to have a major format-lick-ball-editor-sheeple-emulator feel to it. I'm rarely impressed.

All in all, I do visit Slashdot daily, with very low expectations, a grave hatred for that fucking Mega Tokyo anime shit, and annoyed at ads that my inline html proxy/editor doesn*t catch.

I might suggest Ars Technica: The PC enthusiast's resource,The Register * sure it can be crap, but I like it, * the Inquirer and Wired News. Also, for humor, I would strongly recommend Cliff [& Enoch] Yablonski and of course plain Something Awful.

Seriously, I believe a lot of the Slashdot visitation is by trolls, genuinely decent nerdy types (vast minority, even more minute are the ones who are both nerdy and informed/educated/intelligent), and horribly afflicted pre pubescent teens with raging online hormones due to frustrations in real life.

Its gone from bad to worse. The latest was 50 Karma being "Excellent." Is that like, totally, Bill and Ted's Bodacious EXCELLENT adventure or what dude. Lets go watch Anime before we graduate high school. Oh shit, My DBZ fetish got my grades suck, dude. Ill have to go for my GED over my Wind0Z3 XP machine now and pretend im using Linux, man.

Have you ever considered Gobe? It rocked on BeOS, and now its available on the Windows platform. And if you don't want to trust their marketing, then here's a review from Ars Technica. And if you still want to complain, go use vi or emacs or even notepad.

Amigori

The second page of John's Mac OS X 10.1 review contains an index to all of the Ars Technica Mac OS X articles:

http://arstechnica.com/reviews/01q4/macosx-10.1/ma cosx-10.1-1.html