[Dr_Doom]

The host of TechTV's Screensavers reported from Comdex today that the big rumor floating is Apple related. Supposedly AMD will announce in a keynote tomorrow that they will be supplying 64 bit (known as Opturon or Clayhammer) processors to Apple. Maybe the megahertz gap will soon be filled? And, if it happens any time soon they may even beat Intel to the consumer market with 64bit processors. The implications of moving OSX to x86 are mixed, but I suppose the PowerPC chip has dragged them down for long enough.

--------

Sounds super groovy to me. 64 bit goodness at reasonable Hz.

Dr_Doom

[Kaner]

i thought that was yesterday, then i thought it was today (19th), but now its tomorrow?

[Dr_Doom]

You are correct. I forgot to paste the date from the email. The email is from ~10PM last night in regards to today.

[Guy Incognito]

Originally posted by Dr_Doom:
The host of TechTV's Screensavers reported from Comdex today that the big rumor floating is Apple related. Supposedly AMD will announce in a keynote tomorrow that they will be supplying 64 bit (known as Opturon or Clayhammer) processors to Apple. Maybe the megahertz gap will soon be filled? And, if it happens any time soon they may even beat Intel to the consumer market with 64bit processors. The implications of moving OSX to x86 are mixed, but I suppose the PowerPC chip has dragged them down for long enough.

--------

Sounds super groovy to me. 64 bit goodness at reasonable Hz.

Dr_Doom

What would become of app compatibility. 100% of the apps out there would break.

Cocoa apps would all need a recompile and tweaking (which is fine I guess if the app ever gets recompiled at all.) All Carbon apps will have no chance to survive and will have to make their time.�

All Altivec-using apps will be rendered useless or will become slower (but then gain everything back through the higher clockspeed?)

Lets just say it would be total chaos for everyone and everything 'cept maybe servers and admins.

[Dr_Doom]

Servers and admins is exactly the category that I fall into. Must of the applications that I rely on I compile myself anyway, so recompiling would be trivial taken in the context of the performance boost.

I wouldn't be suprised if most developers felt the same way.

[BuonRotto]

I thought that since apps worked on top of the kernel, that only minor recompiles would be necessary, and that a new kernel would do most of the work in the transition?

[Millennium]

We can only hope that this is not true.

But to be honest, I don't think it's likely anyway. And for the last time, it's not PPC which has been holding Apple back; it's Motorola. Get a good chipmaker on the architecture, and you'll see the gap vanish, while maintaining the superior PPC architecture..

[pliny]

Carbon apps could be recompiled for mach-o--so they can be ported to x86 too. This would be a fresh burden though to developers who instead have focused on including CFM compatibility in their Carbon apps. Which is most of them--anyone know which developers have built Carbon apps without CFM code in favor of mach-o?

[LightWaver-67]

*-*-*-* DISCLAIMER *-*-*-*

I am in NO WAY a technical genius... I only know bits and pieces of what I read online... and even that... i only PRETEND to understand. Having said that:

___________________________________

I'm sure this has been discussed before, but if Apple decides to migrate to a new chipset... wouldn't one of two things need to happen...:

1) They would need it to run on some sort of abstraction layer to allow it to make calls to the chip (or something along those lines) that would make it potentially slower or less stable... or BOTH.

or

2) Developers and programmers would need to (yet again) modify or totally re-write code to comply with such modifications.

Now... I don't KNOW any of it to be true... I know about as much about processors and computer code, as I do about nuclear fission or brain surgery.

Bottom-line... I REALLY hope they don't need to go back to developers and 'sell-them' on the idea of re-coding anything... or having to say to users... "Hey.. look how fast our processor is now... even though it gets slowed-down due to emulation and is less stable..."

- crossing fingers -

[Guy Incognito]

Originally posted by pliny:
Carbon apps could be recompiled for mach-o--so they can be ported to x86 too. This would be a fresh burden though to developers who instead have focused on including CFM compatibility in their Carbon apps. Which is most of them--anyone know which developers have built Carbon apps without CFM code in favor of mach-o?

Tell that to Adobe and Microslap.

[Jeff Jones]

If they would do that -- which I don't believe at all -- then only for the server-side. Assuming that they have the OS working on x86, there is no way in hell that the likes of Adobe (and Apple for that matter) will port their apps to Intel.

AMD has shown that is is able to simulate a PPC CPU, but I don't think this is ready for prime time yet. And if it were, it would most probably be slow.

And just to confuse you: This could not be the reason why Quark is so reluctant to release XPress 6. Or could it?

Edited: typ-o

[pliny]

Originally posted by Guy Incognito:


Tell that to Adobe and Microslap.

?
I don't think they would look on it too kindly. It can be done, but it would put a new burden on developers who've only recently rewritten their apps. That's why I don't think a move to x86 is likely or even desirable (my point).

[cwasko]

I'm not going to be so bold as to say that Apple will never use an AMD chip (in a Macintosh); however, I'll go out on a limb to say it ain't gonna be announced tomorrow (for the Macintosh).

Thw whole Apple world is still in a big time transition from OS9 to OSX. Changing the CPU right now will just plain put them out of business. No one will buy the new machines, because the only software it would run would be OSX. No developer would make any apps for them, becasue no one has the machines. Apples stock plumets, profits turn into loss. Apple dies. This is obviously not something Apple wants to happen, nor myself.

[Dr_Doom]

If you take the consumer business out, and transition the Opteron into the very young server market that Apple is trying to break into, you can see that transitioning _that_ user base to an x86 architecture would have a minimal impact since the quantity of deployed chassis are still quite small.

Those server users IMO would be more likely and able to recompile their *nix apps and get on down the road.

I believe it would be foolish for Apple to lump the consumer base and the business base into the same basket in terms of hardware and developement since they are such fundamentally different markets.

I mean, I don't have an XServer sitting on my desk.


Er . . . . yet


Dr_Doom

[Chuckit]

Originally posted by cwasko:
I'm not going to be so bold as to say that Apple will never use an AMD chip (in a Macintosh); however, I'll go out on a limb to say it ain't gonna be announced tomorrow (for the Macintosh).

Thw whole Apple world is still in a big time transition from OS9 to OSX. Changing the CPU right now will just plain put them out of business. No one will buy the new machines, because the only software it would run would be OSX. No developer would make any apps for them, becasue no one has the machines. Apples stock plumets, profits turn into loss. Apple dies. This is obviously not something Apple wants to happen, nor myself.

Hasn't Apple already announced that their new line of machines will boot only into OS X? I thought that was where this whole rumor came from in the first place.

[absmiths]

Originally posted by Dr_Doom:
If you take the consumer business out, and transition the Opteron into the very young server market that Apple is trying to break into, you can see that transitioning _that_ user base to an x86 architecture would have a minimal impact since the quantity of deployed chassis are still quite small.

Those server users IMO would be more likely and able to recompile their *nix apps and get on down the road.

I believe it would be foolish for Apple to lump the consumer base and the business base into the same basket in terms of hardware and developement since they are such fundamentally different markets.

I mean, I don't have an XServer sitting on my desk.


Er . . . . yet


Dr_Doom

Microsoft already tried this with Windows NT running on different processors - and the idea flopped. Apple doesn't want to have to release all software that *might* run on an XServe in PPC/x86 dual architecture, any more than any other company would want to have to choose between the two markets explicitly by compiling for PPC or x86 or some FAT binary.

Instead they should focus on greater parallelism and better chips (in that order - I would rather have 10 500MHz chips in a server than a single 5GHz chip).

[mitchell_pgh]

As I have posted elsewhere...

Apple could come out with OS X for x86 in a server version. That would shut everyone up... Joe Blow won't care because he loves his Mac, IT guy might care because OS X Server is easy to maintain...

Serioulsy, if you got Office, Adobe, and Apple on board with an x86 port, that would account for 50% of the Mac market...

I just want Apple to make more $$$ and give M$ a run for their money...

[mitchell_pgh]

Originally posted by absmiths:

Instead they should focus on greater parallelism and better chips (in that order - I would rather have 10 500MHz chips in a server than a single 5GHz chip).

That statement reminds me of when I was watching a BeOS seminar at Carnegie Mellon University...

They said that Processor speed would become a thing of the past because dual and quad configurations would prove to be more cost effective. They had two similarly priced systems running (one with a high end 604 chip, and another with two low end chips)...

The performance was considerably better for a majority of applications on the dual system. Even look at P4 systems... two low end P4s equal about the same price as one high end P4...

Case and Poing:
$418 - Xeon 2.6GHz
$245 - Xeon 2.4GHz
$245 _-_ Xeon 2.2GHz
$198 _-_ Xeon 2.0GHz

Tell me how one 2.6 GHz Xeon is going to blow away a dual 2.0 Xeon system (and the dual system would be $20 cheaper!)

I know the board would cost more, but I think you get my point...

[Millennium]

Originally posted by Chuckit:
Hasn't Apple already announced that their new line of machines will boot only into OS X? I thought that was where this whole rumor came from in the first place.

Yes, but they also said that Classic would still work. That means there has to still be a PPC chip, because a PPC can't be emulated well on an X86.

However, it is possible, now that I think about it, that Apple may use AMD's new HyperTransport chipset. This doesn't change the PPC architecture at all, but it should very neatly solve several of the other problems currently plaguing Mac hardware, notably bus speed.

[alex_kac]

Originally posted by Guy Incognito:


What would become of app compatibility. 100% of the apps out there would break.

Cocoa apps would all need a recompile and tweaking (which is fine I guess if the app ever gets recompiled at all.) All Carbon apps will have no chance to survive and will have to make their time.�

Carbon would work just fine. Cocoa would work just fine. Basically recompile and tweaking. Carbon is NOT the old Mac OS toolbox. CLASSIC would not work fine.

[cpac]

Originally posted by Millennium:
a PPC can't be emulated well on an X86.

Honest question from an ignorant:

PPC basically = RISC = Reduced Instruction Set Chip

x86 basically = CISC = Complete Instruction Set Chip

I understand that the trade off (from x86 to PPC) was to eliminate some of the rarely called instructions and that by having to worry about fewer instructions overall, the chip could be faster.

Question though: Doesn't the idea of CISC imply that it comprehends RISC? That is, aren't all RISC instructions included in CISC?

If so, why would emulation be so difficult?

(I realize this is a very dumbed-down version of things, but that's all some of us know).

[krove]

Originally posted by mitchell_pgh:
As I have posted elsewhere...

Apple could come out with OS X for x86 in a server version. That would shut everyone up... Joe Blow won't care because he loves his Mac, IT guy might care because OS X Server is easy to maintain...

Serioulsy, if you got Office, Adobe, and Apple on board with an x86 port, that would account for 50% of the Mac market...

I just want Apple to make more $$$ and give M$ a run for their money...

What good is an OS X server if it can't netboot (i.e. running apps on the server for the client akin to a terminal server)? Nothing!

Isn't the keynote over by now?

[brainchild2b]

First of all many of you are thinking so very narrow minded.

*CEO of AMD was formally at Motorola in semiconductor department.

*Velocity Engine (aka AltiVec) Apple is rumored to have purchased the rights too.

*The new AMD chip in theory could emulation PPC nicely.

*AMD could make a powerpc emulation type chip and put it on the board with it's new 64 bit chip.

*Apple could make the transition nearly seemless.

*Apple obviously has been working thinking about the AMD 64bit for quite some time, you can pick that up through it's darwin mailing list or stuff mentioned by AMD employees.

*Apple is commited to hypertransport

*AMD & Apple have always had a strong relationship

*The New AMD chip has been delayed for a while, for reasons other than the chip being ready.

*Apple went from 68k (and entirely new kind of instruction set) to RISC fairly seemlessly, and could have been cooking up ways to make everything works seemlessly on AMD for the last 4 years.

*Apple CEO Steve Jobs publicly stated that Apple was considering all platforms and that x86 was an option.

*AMD could put Altivec on setup with new hammer chip for apple.

*Strange how apple's own mailing list (darwin) refers to Athlon-XP as the "offical" name of the chip when AMD claims it hasn't decided yet.

*Darwin is almost ready and has documentation as of recently on running on x86, athlon 64bit. Which is funny for a chip that's not out yet, and since when has Apple been in a hurry to update darwin for unreleased processors? (usually they take their sweet time)

*Apple could have a customized AMD box that ran OS X, OS X would NOT run on any x86 machine only Apple built AMD machine.

*It makes good business sense, keep pace with wintel world.

*IBM's new chip is expensive, will top out faster than wintel (AKA motorola fiasco all over again)

*I saw OS X DP run on AMD processor 2 years ago, and everyone said I was full of crap back then, funny how the rumors are now don't u think?

I'm not saying that it is for sure, everything should be taken with a grain of salt. I just want people to realize it's alot more doable than you think.

[clarkgoble]

From what I understand over on the Darwin boards they say that Darwin x86 doesn't run terribly well. It needs quite a few optimizations to match the PPC. If Apple was planning on coming out with a x86 version of OSX, one would expect that more work would be being done by the Darwin team on those optimizations.

Further there is the oft mentioned problem of interpreting PPC code on a x86. I brought this up on the hardware boards and other than wishful thinking I've heard no response to this.

[krove]

Originally posted by brainchild2b:
First of all many of you are thinking so very narrow minded.

*CEO of AMD was formally at Motorola in semiconductor department.

*Velocity Engine (aka AltiVec) Apple is rumored to have purchased the rights too.

*The new AMD chip in theory could emulation PPC nicely.

*AMD could make a powerpc emulation type chip and put it on the board with it's new 64 bit chip.

*Apple could make the transition nearly seemless.

*Apple obviously has been working thinking about the AMD 64bit for quite some time, you can pick that up through it's darwin mailing list or stuff mentioned by AMD employees.

*Apple is commited to hypertransport

*AMD & Apple have always had a strong relationship

*The New AMD chip has been delayed for a while, for reasons other than the chip being ready.

*Apple went from 68k (and entirely new kind of instruction set) to RISC fairly seemlessly, and could have been cooking up ways to make everything works seemlessly on AMD for the last 4 years.

*Apple CEO Steve Jobs publicly stated that Apple was considering all platforms and that x86 was an option.

*AMD could put Altivec on setup with new hammer chip for apple.

*Strange how apple's own mailing list (darwin) refers to Athlon-XP as the "offical" name of the chip when AMD claims it hasn't decided yet.

*Darwin is almost ready and has documentation as of recently on running on x86, athlon 64bit. Which is funny for a chip that's not out yet, and since when has Apple been in a hurry to update darwin for unreleased processors? (usually they take their sweet time)

*Apple could have a customized AMD box that ran OS X, OS X would NOT run on any x86 machine only Apple built AMD machine.

*It makes good business sense, keep pace with wintel world.

*IBM's new chip is expensive, will top out faster than wintel (AKA motorola fiasco all over again)

*I saw OS X DP run on AMD processor 2 years ago, and everyone said I was full of crap back then, funny how the rumors are now don't u think?

I'm not saying that it is for sure, everything should be taken with a grain of salt. I just want people to realize it's alot more doable than you think.

Hmm...all heresay and rumors. Narrowminded. If someone told you than man had already landed on Mars, would you believe them? Just because its doable doesn't mean its been done.

We're not be narrowminded: there is no proof either way, and you never addressed the Classic issue. Skeptecism keeps everything in perspective.

[Meteo]

euh... As far as I could remember, Athlon are NOT CISC processors but RISC. so call them x86 if you want, they are not, they emulate x86 I think.
that is why AMD chips have lower frequency than P4 to match the P4 perfs.
It could be great if a big and creative and dynamic company as AMD is make processors for AAPL, insted of that f**king company called M*t*r*l*
IBM new chip is expected for the end of 2003 and it would certainly be too late (for us...). I mean intel (and AMD!)will certainly have launch more powerful x86 chips on the market that can "easily" beat G4 (even bi proc). But, it seems obvious that choose IBM is certainly more reliable...

[Nebrie]

Originally posted by brainchild2b:
[B]First of all many of you are thinking so very narrow minded.

*Velocity Engine (aka AltiVec) Apple is rumored to have purchased the rights too.

MOSR said this. That's all that needs to be said.

*Apple CEO Steve Jobs publicly stated that Apple was considering all platforms and that x86 was an option.

Reuters claimed that Steve Jobs had directly told this. Later we found out that Steve Jobs had directly told Reuters this... in a room with dozens of other reporters, except that the x86 part never happened and the IBM part which Reuters ommited, did. (I guess they assumed IBM instantly meant IBM compatible).

*Strange how apple's own mailing list (darwin) refers to Athlon-XP as the "offical" name of the chip when AMD claims it hasn't decided yet.

Well duh. They have the name but they want to keep their options open until it's time. Almost every major company does this. It's not rare.

*It makes good business sense, keep pace with wintel world.

Or hacked so OS X can run on any PC and completely destroy Apple financially. That's what I'm worried about.

*IBM's new chip is expensive, will top out faster than wintel (AKA motorola fiasco all over again)

Yeah, I never got why everyone kept pushing Apple to dump Moto over IBM (in the pre 970 era). IBM was slower and more expensive. I guess they just seemed more cool.

*I saw OS X DP run on AMD processor 2 years ago, and everyone said I was full of crap back then, funny how the rumors are now don't u think?

I also saw a band of monkeys marching down my street today, so?

[JLL]

Originally posted by brainchild2b:
*IBM's new chip is expensive, will top out faster than wintel (AKA motorola fiasco all over again)

Says who? PowerPCs are usually cheaper than x86 processors.

[godzookie2k]

Originally posted by JLL:


Says who? PowerPCs are usually cheaper than x86 processors.

uhm, actually no.

[Shuh]

Dr_Doom said:

The host of TechTV's Screensavers reported from Comdex today that the big rumor floating is Apple related. Supposedly AMD will announce in a keynote tomorrow that they will be supplying 64 bit (known as Opturon or Clayhammer) processors to Apple. Maybe the megahertz gap will soon be filled? And, if it happens any time soon they may even beat Intel to the consumer market with 64bit processors. The implications of moving OSX to x86 are mixed, but I suppose the PowerPC chip has dragged them down for long enough.

These rumors have been promulgated by people who are ignorant of the fact that the PPC architecture already has its own 64-bit chip on the horizon, the 970/GPUL. Apple doesn't _need_ AMD to get faster 64-bit consumer desktop chips. So this is a non-issue.

[MacGorilla]

I also saw a band of monkeys marching down my street today, so?

Really? send 'em my way. Its boring around here.

[Brass]

Any reason why AMD couldn't produce PPC chips for Apple?

Apart from licensing, which I assume Apple could arrange.

[King Bob On The Cob]

Does no one think that maybe, just maybe Apple may not have the same companies making chips for it's entire line anymore? AMD making the PowerMac line, Moto making the (i)Consumer line, IBM making the Server line. anyone want to comment on this?

[Terri]

My 2� says AMD PowerPC chip.

Mac X has all the Mac people and also many people from other platforms like linux. With a much lager potential user base it makes it worthwhile to make a chip just for Mac X.

Apple still controls the chips through licenses with AMD so if someone else wants to make a box that runs Mac X Apple could license the chips much they way the Wintell world works now, but they would be selling a license both for the hardware technology and the software technology.

Then again I may see flying pigs outside my window in the morning�

[K++]

Originally posted by Terri:
My 2� says AMD PowerPC chip.

Mac X has all the Mac people and also many people from other platforms like linux. With a much lager potential user base it makes it worthwhile to make a chip just for Mac X.

Apple still controls the chips through licenses with AMD so if someone else wants to make a box that runs Mac X Apple could license the chips much they way the Wintell world works now, but they would be selling a license both for the hardware technology and the software technology.

Then again I may see flying pigs outside my window in the morning�

I completely agree, it would such an asinine move for Apple to move to x86 now considering that the x86 has exhausted its life cycle and should soon die, most x86 chips are mixed RISC, CISC anyway nowadays, I know AMD's are. So why not take out the x86 emulation and make a pure PowerPC 64bit chip for Apple. It would be nice but I am highly skeptical, and its about 8 PM now and I have heard nothing of this anywhere online.

[sadie]

It would make sense to have more than one company producing PPC processors at once. That way, when motorola (for example) can't fulfill the demand for a certain line, there's somebody else happy to fill in the gap with an equivalent. Moto's niche monopoly is probably a large part of the reason Apple's had so much trouble with them.

It also fits in with the trend of increasing number of dual and quad systems.

But this specific announcement, I expect will just be HyperTransport.

[Camelot]

Originally posted by King Bob On The Cob:
Does no one think that maybe, just maybe Apple may not have the same companies making chips for it's entire line anymore? AMD making the PowerMac line, Moto making the (i)Consumer line, IBM making the Server line. anyone want to comment on this?

The trouble with this idea is quantity.

You always, in any business, get a better price break by buying in quantity.

If you're buying, say, 500,000 processors from AMD, 1 million processors from Motorola, and 500,000 processors from IBM, you'll be paying a LOT more than if you gave any one company an order for 2 million chips.

[Millennium]

Originally posted by cpac:
Honest question from an ignorant:

PPC basically = RISC = Reduced Instruction Set Chip

x86 basically = CISC = Complete Instruction Set Chip

I understand that the trade off (from x86 to PPC) was to eliminate some of the rarely called instructions and that by having to worry about fewer instructions overall, the chip could be faster.

Question though: Doesn't the idea of CISC imply that it comprehends RISC? That is, aren't all RISC instructions included in CISC?

If so, why would emulation be so difficult?

(I realize this is a very dumbed-down version of things, but that's all some of us know).

One of the biggest reasons you can't emulate a PPC well on x86 well has to do with something called registers. These are tiny chunks of superfast memory which are inside the processor. Usually they're used to hold either very small but important pieces of data, or pointers to data in main memory.

The PowerPC has 64 of these registers: 32 for data, and 32 for instructions. The x86 architecture provides for only eight, and four of these can't be used reliably at all, and three of the others are apt to change when certain instructions are run.

When you have to emulate eight registers and have 64 real registers, your task is easy; just reserve right real registers. But when you have to emulate 64 registers with only eight, you can't do that. You have to store things in regular memory, which is a massive slowdown. Not just because main memory is slower than register memory, either; instead of just reading a value from a register, you have to save the old value of the register you're using somewhere in main memory, move the value of the register you want to emulate into the register you want to use, and then read the register. Even if main memory were as fast as registers, that would at least triple the time it takes to read an emulated register, and in reality it's much, much slower than that.

This wouldn't be so bad, except that there's an interesting side effect of the PPC having so many registers: they get used. Programmers and compilers take heavy advantage of the register-richness of the architecture; it's one of the major ways to optimize an app for the PowerPC chip. This only slows things down even more, though, when you're emulating a PPC chip on a machine that doesn't have enough registers for the job. And it should be noted that neither IA-64 not x86-64 have all that many more registers (x86-64 has twelve; I don't know about IA-64), so these architectures still won't be able to emulate PPC well.

As for instruction sets, here's the thing. Although it's true that the PowerPC has fewer instructions than x86, that doesn't mean there's a one-to-one mapping between them. I mean, a mov instruction (move the contents of a register into main memory, or vice versa) is pretty similar for both. But eieio or stfsux? Those are both real PPC instructions. Come to think about it, what do those instructions do anyway, other than illustrate the PPC design team's penchant for humor in technology? Which, incidentally, can also be seen right at the beginning of any PEF application; th first four bytes spell out "Joy!" (I believe it's meant as a Ren and Stimpy reference, kinda like stfsux is a Simpsons reference).

[cpac]

Originally posted by Millennium:

...detailed explanation...

Thanks very much!

[passmaster16]

Originally posted by Brass:
Any reason why AMD couldn't produce PPC chips for Apple?

Apart from licensing, which I assume Apple could arrange.

I'm not sure AMD would be willing to dump the R&D into producing a chip for a company that only has 3% of the market. I think Apple is in a situation where they can only use whats available to them. IBM and Motorola use their chips in other products as well as Apple's otherwise I'm not sure if it would be worth it for them to produce those chips.

[JLL]

Originally posted by godzookie2k:



uhm, actually no.

Actually yes, I'm not talking about the price of one processor that the average Joe can buy.

In quantities of 1,000 or more, the price of a PowerPC is much lower.

Price of 1GHz G4: $296

2.8GHz Pentium 4: $401

2.6GHz: $305

[Developer]

Originally posted by passmaster16:

I'm not sure AMD would be willing to dump the R&D into producing a chip for a company that only has 3% of the market. I think Apple is in a situation where they can only use whats available to them. IBM and Motorola use their chips in other products as well as Apple's otherwise I'm not sure if it would be worth it for them to produce those chips.

Wouldn't that depend on how much additional R&D it would cost to make the processors handle a different instruction set? I thought the AMD processors are doing the x86 instruction set via emulation anyway.
So if it would cost 3% additional R&D to make them handle the PowerPC instruction set, it might be worth it.

[clarkgoble]

While the Hammer has only 16 real registers and then 48 virtual registers, the Intel equivalent has 128 registers. There may be some restrictions on some of the Intel registers though.

[Nonsuch]

Originally posted by Millennium:
kinda like stfsux is a Simpsons reference).

Yeah, stuff sucks.