Mac OS X Intel Build 8F1111

[displaced]

I'm starting to get excited about the intel Apples. At first I never would have thought of buying one (for the first few years), but it sounds like Apple is working hard on making it "flawless." They know they can't blow it this time (Apple has a bad history of this).

Rosetta sounds cool also.

I'd have said Apple's track record was pretty good for this sort of thing. The 68K to PowerPC transition wasn't too bad, all things considered. And Classic support in OS X was an elegant solution to a thorny problem. The move to OS X itself was pretty astonishing, completely altering perceptions of the entire platform in a little more than 4 years (personally, I'd argue a little less than 3 years!)

 

[aafuss1]

Nice to see a Intel-ready version of iTunes. I wonder if the TPM code has been tightend up even more.

 

[EricNau]

I'd have said Apple's track record was pretty good for this sort of thing. The 68K to PowerPC transition wasn't too bad, all things considered. And Classic support in OS X was an elegant solution to a thorny problem. The move to OS X itself was pretty astonishing, completely altering perceptions of the entire platform in a little more than 4 years (personally, I'd argue a little less than 3 years!)

I should have been clearer. Apple is usually good about creating good computers, but if I recall correctly, their first models after a big switch have been lacking in features, which in the past has cost them, big time. Many say this is why Apple didn't "boom" like Windows because Apple's first computers were lacking in features.
I think after the switch to PowerPC their computers were lacking things like USB and such. (this is coming from someone who has one, can't attach anything to it )

 

[nagromme]

I'm wondering what is the possibility that the dual core Yonah running apps in Rosetta could at least keep up, if not match, apps running natively on the 1.67Ghz PPC PowerBooks.

I think it's a sure bet that Rosetta will outrun a G4 in many situations. Especially when it comes to perceived responsiveness (as opposed to the duration of lengthy renders and conversions).

I say this because of comments from game developers at InsideMacGames about some 3D games performing very well under Rosetta. Well enough that there's no NEED to port some of them. And that's on the dev kits, which are pretty ordinary single core machines.

 

[nagromme]

AltiVec means Mac games will work. w00t!

Note that many games that say they require a G4 do NOT require AltiVec. A moot point, possibly, but Quake 3 engine games, for instance don't use AltiVec and reportedly run great even in the OLD Rosetta. Despite their requirements stating a G4. It turns out that stating a G4 is often done just because no G3 Mac was made fast enough. Not because the game--or other app--is incompatible with a G3.

 

[nagromme]

I think after the switch to PowerPC their computers were lacking things like USB and such. (this is coming from someone who has one, can't attach anything to it )

USB wasn't even around when they went PowerPC--that was in the mid-90s. But Apple (with the iMac) was one of the first, if not THE first, PC makers to widely adopt USB.

If you mean that when a new OS or CPU architecture comes out, it takes time for apps to fully take best advantage of it, that's true. Luckily, that process has been long since underway.

(Sorry for the triple post... 8 minutes and nobody but me is posting? )

 

[EricNau]

USB wasn't even around when they went PowerPC--that was in the mid-90s. But Apple (with the iMac) was one of the first, if not THE first, PC makers to widely adopt USB.

If you mean that when a new OS or CPU architecture comes out, it takes time for apps to fully take best advantage of it, that's true. Luckily, that process has been long since underway.

(Sorry for the triple post... 8 minutes and nobody but me is posting? )

USB and firewire came out in 1996 - not entirely sure when the Power Mac G3 All-in-one came out, but that's te computer he was refering to. I would be surprised if he didn't know what he was talking about, he really knows a lot about Mac Computers. Maybe I got a detail wrong.

I just know he said he would never buy a intel Mac right away because of his past experiences with Apple.

 

[GregA2]

I'm glad to see they added altivec support to Rosetta... Do you suppose FCP would work decently on a system like that?

 

[nagromme]

Maybe I got a detail wrong.

He said "when they went PowerPC", not "when they went G3"--maybe that's the missing detail

Still, Apple was ahead of other makers when it came to introducing USB.

Of course, the later the model of anything, the better. Waiting--if you can--always has benefits!

 

[Thomas Harte]

I've always wondered just how effective Altivec is though. I've always thought it was more of a marketing term rather than an engineering one.

To add to what has already been said, not only is Altivec a real engineering term, but it is a feature that is likely to be comparitively less speedy in Rosetta (in terms of "how many MHz is my Rosetta CPU?") because it doesn't map very well to the Intel solutions. At least with SSE3 Rosetta only has to deal with a 50% drop in registers, but whereas Altivec does most things in three address format (i.e. ra = rb + rc) the various Intel solutions do things in two address format (i.e. rb = rb + rc) and the permutation features of Altivec seem to be entirely absent from the Intel chips, although they're mostly used for formatting the result of Altivec computations so some additional costs there aren't necessarily crippling.

Of course the Intel architecture does things that would map badly to an emulated PowerPC, but that's completely beside the point. Universal binaries will almost certainly get more vector throughput from an Intel than a PowerPC.

I'm starting to get excited about the intel Apples. At first I never would have thought of buying one (for the first few years), but it sounds like Apple is working hard on making it "flawless." They know they can't blow it this time (Apple has a bad history of this).

But we have to factor in Apple's "secret agenda" here. Darwin is open source and has been buildable and usable on Intel hardware for many years. The keynote announcing the transition was performed with Intel hardware, implying that they probably could have put out a non-"flawless" first generation Intel box immediately. And unlike the transition to PowerPC, the Intel solutions already have a long history of deployment in exactly the kind of machines Apple will produce, aesthetics and software aside, so the knowledge and expertise is already widely available.

 

[Dorkus Maximus]

I'm glad to see they added altivec support to Rosetta... Do you suppose FCP would work decently on a system like that?

Good question. FCP is definitely an Altivec intense application to say the least. I'm not sure how it's progressed since its move to the G5, but I know I would be "porked" to say the least without Altivec on my Powerbook. I'm guessing that the move to the G5 may have helped with this transition, since it doesn't rely on the Altivec engine (I may be wrong).

Regardless I'll be picking up a dual-core powerbook the day they're released. It's unfortunate that the first systems will be problematic and outdated quickly (on the hardware end) as Apple quickly learns what doesn't work and what does with a few million beta testers. I have tons of faith in Apple, and think they'll do a better job than anyone else ever could at such a transition... but life is life. There are going to be some very fun, and by fun I mean horrible, growing pains.

Lastly, it would've been a fun idea to sell beta intel machines to users who want 'em, or even rent them out. I'd fork out a $1,000 for one or pay a few quid a month.

 

[mdavey]

Can we just skip December and go straight to MWSF?

I suggest we have a poll about that. I would be in favour. I can see many benefits, including having more money available for MWSF (saving on all those xmas present purcahses - even taking into account loss of one month's salary).

Perhaps we should have an option to swap the order of December and January? Might be popular with some.

 

[generik]

Regardless I'll be picking up a dual-core powerbook the day they're released. It's unfortunate that the first systems will be problematic and outdated quickly (on the hardware end) as Apple quickly learns what doesn't work and what does with a few million beta testers. I have tons of faith in Apple, and think they'll do a better job than anyone else ever could at such a transition... but life is life. There are going to be some very fun, and by fun I mean horrible, growing pains.

Is FCP a SMP compatible application?

 

[gnasher729]

Well, if Rosetta supports Altivec, then there is no reason for me not to get the first Intel Powerbook. I think not having Rosetta support Altivec would deter many, many users from buying Intel Macs when they first arrive.

Having Rosetta _support_ Altivec would actually deter me from ever buying an Intel Mac. Emulating a vec_perm instruction takes sixteen loads and stores, and in good have Altivec code half the instructions will be vector permutes. Expect emulated Altivec code to run much slower than emulated G3 code.

 

[generik]

Having Rosetta _support_ Altivec would actually deter me from ever buying an Intel Mac. Emulating a vec_perm instruction takes sixteen loads and stores, and in good have Altivec code half the instructions will be vector permutes. Expect emulated Altivec code to run much slower than emulated G3 code.

Except processing power will probably scale much faster on the x86 front than PPC. With dual core *mobile* chips, I really see a good future for Intel Macs.

 

[sluthy]

Because it CAN emulate Altivec code (Altivec being the engineering term "Velocity Engine" being the marketing term I remember), does that mean non-Altivec code HAS to be emulated the same way? If not, then why is there any reason for Altivec compatibility to be a deterrant?

Whatever - for me, any modern Intel chip could outdo my 500MHz G3 .

 

[aegisdesign]

What good is a 1.6Ghz CPU and RAM running at 333MHz when the FSB runs at 167Mhz? Like trying to drive water from a firehose through a straw.

Erm, you presumably weren't aware that the Intrepid2 controller in the new Powerbooks runs at a full 333Mhz then?

Not everything has to go through the FSB. A lot of the time Intrepid2 will shove data from RAM, GPU, Network or the harddisk straight into RAM via DMA at a full 333Mhz without involving the FSB at all. ie. at exactly the same speed as a Pentium-M with PC2700 DDR2 RAM.

FSB speed isn't as important as you imagine.

 

[mad jew]

Altivec is an engineering term. I was saying slightly tongue in cheek that it's more of a marketing term because Apple would have you believe it's more useful than it actually is.

 

[aegisdesign]

Having Rosetta _support_ Altivec would actually deter me from ever buying an Intel Mac. Emulating a vec_perm instruction takes sixteen loads and stores, and in good have Altivec code half the instructions will be vector permutes. Expect emulated Altivec code to run much slower than emulated G3 code.

That's what I was wondering too. It's fine and dandy saying it now 'supports' AltiVec but how fast will it really be when SSE is slower than Altivec anyway and doesn't support all the features in AltiVec. Dropping back to pretending to be a G3 may be better.

Still, if it now runs code that otherwise would just refuse to run then that's perhaps a bonus.

 

[runninmac]

Can we just skip December and go straight to MWSF?

No honestly can we?

MWSF is going to be like X-mas all over again. I only need 1 x-mas a year why not have it MWSF?

 

[aegisdesign]

Altivec is an engineering term. I was saying slightly tongue in cheek that it's more of a marketing term because Apple would have you believe it's more useful than it actually is.

Bull.

I've both G3 and G4 machines running at the same Mhz. iTunes encoding, video transcoding and the whole UI runs many times faster on the AltiVec equipped G4s.

It's the reason my G5 1.8 is almost twice as fast as my Pentium-M 1.7 at some media transcodes.

 

[aegisdesign]

Is FCP a SMP compatible application?

Of course. It's even capable of splitting tasks out to other Macs on the network.

 

[mad jew]

I've both G3 and G4 machines running at the same Mhz. iTunes encoding, video transcoding and the whole UI runs many times faster on the AltiVec equipped G4s.

Weird. My experiences are quite different. Maybe it's just how I use my Macs but in the limited time I used a 700MHz G3 iBook, I didn't notice all that much of a difference with my current 800MHz G4 iBook for CPU-related tasks. The G4 has more RAM so it was snappier in that sense but I was quite taken aback at how useless AltiVec seems to be.

It's the reason my G5 1.8 is almost twice as fast as my Pentium-M 1.7 at some media transcodes.

And AltiVec is the only difference between those two processors.

Let's compare Apples with Apples.

 

[aegisdesign]

Regardless I'll be picking up a dual-core powerbook the day they're released.

Unless Apple also release an native Intel FCP on the same day then FCP running on your old PowerPC powerbook would most likely be faster and I'd guess by some margin. If they get even 10 to 20% speed in AltiVec emulation I'd be surprised.

Bits may run a bit quicker though now that they have Quicktime 7.0.4 as intel native. I'm not sure however how much in FCP is Quicktime and how much is their own code.

 

[aegisdesign]

And AltiVec is the only difference between those two processors.

Let's compare Apples with Apples.

My iMac G5 1.8Ghz has a 600Mhz FSB, 512KB L2 cache, 1GB PC3200 DDR RAM

My PentiumM 1.7Ghz has a 533Mhz FSB, 2MB L2 cache, 1GB PC4200 DDR2 RAM

They're comparable most of the time. Sometimes the iMac is faster, sometimes the PentiumM. Sometimes the difference is huge - eg. OpenGL is hugely faster on Windows, anything media related is normally faster on the Mac.