First Intel Macs on track for January - Page 5

Which future Intel CPUs are dual-processor friendly (besides the Xeon)?

What Intel quad (dual-dualcore) would likely have similar performance as the G5 2.5 GHz quad? I am trying to gauge how long until the MacIntel powermac will take to catch up to the present quad. Any guidance would be appreciated.

Xeon is it, and i don't know if they're dual core yet. But i have heard that Opteron > Xeon and PPC970 > Opteron.

You make very good points, Mr. DCQ.

I too am afraid of OS-X on any old box out there; Macintosh is not the OS, and it is not the machines... it is the whole seamless shebang, a shebang that could turn very easily into a fizzle if prying fingers are allowed into the mix.

The problem with the dual-booting on the machines (and aren't there viruses that attack intel BIOSes as well?), is that it too can taint the Apple experience. Even if someone installs windows and gets their machine burned by malware; it will definitely take the bloom off the rose... or the sheen off the Apple.

Another quick question: Is intel's hardware DRM going to make it into macs?

Hope springs eternal,

Mandricard
AppleOutsider

So after reading through this thread it now seems to me that Apple will do their own thing. They haven't made any statements that would preclude this. Sure the new Macs will have some as yet unknown Intel proc in it, but really why would Apple even bother trying to build a box that worked internally like typical Wintel systems? Those system are full of legacy compromises that Apple can ignore. Completely. Some of the stuff the tech folk have discussed here is intriguing. The different cpu's, the busses, the memory controllers, EFI etc.

They can do anything they want to make it work best for their OS. Everybody else is on their own. Yes I know the dev boxes they sent out are standard stuff, but it doesn't have to remain that way.

This is exciting.

Hey sunil, I see you moved. Housing situation a problem?

Hello people!

How many times have I seen this already?

does anyone read the news?

Apple is patenting a security method for this. Even if you somehow manage to get the OS to run on another machine, you won't be able to run programs on it

It's not in effect on the developers machines yet, but by release it will be ready.

I thought Leopard was going to be introduced with Intel Macs... now it's just an upgrade again?

We have no idea. We just like to talk.

And if there's enough interest, it wil be cracked in 5 minutes - just like various Microsoft anti-piracy/licencing schemes, MPAA schemes, and most other copy protections, OS licence protections, protections to applications, et al.

You sound like you actually believe that a piece of software will be impossible to crack.

I see, well this sounds good none the least!

Yeah, I do believe that the almost required idea that all security will be cracked in five minutes by some pimply kid panting in his room is wishful thinking.

Everyone remembers the few successful cracks, but don't know about or remember the ones that weren't.

DVD security was cracked by that kid Jon.

What's forgotten is that it was only cracked because the manufacturer of the disks forgot to remove the public key from the software before pressing the disks.

It's never been cracked again.

iTunes was cracked when it first came out, fixed, cracked, fixed again, and hasn't been cracked since. There were a couple of bugs. Once they were fixed, that ended it.

It's agreed that the new hi-def disks coming out will be uncrackable, that even if somehow one disk was cracked, it would only work for that release, but that it's unlikely to happen.

Apple's patents contain well understood methods of code substitution as well as other methods combined with a hardware chip. Apple hasn't instituted the patent yet, but it has much weaker security just based on the chip. Developers say that the 10.4.3 release contains much more secure protection than earlier releases, even though it doesn't use Apple's patented methods.

Apple's patent has been looked at by security experts who have said that while it could be possible to get the OS to work on another PC, critical parts of the OS would be missing as they reside in the chip. due to the code substitution methods, which code it would be could be impossible to determine, as it would be different on each copy of the OS.

Therefore, while the OS MIGHT run, getting apps to run on it could be impossible.

Of course we'll see after it comes out, but it does look difficult.

Remember that when security is built-in from the beginning it's much more effective then when it's built on top, later.

XP's security that they recently put in doesn't work well because it's an add-on. Once you're through it, you're done.

Um, Leopard is an upgrade the same way Tiger, Panther and Jaguar were, with the only difference that it'll run on two architectures instead of just one.

also vmware for linux and windoze

One question I have for all of you worrying that running Windows on Macs would actually be could be the death of Apple. Won't the developers go where the market is? And if the market exists for SW running on the Mac OS, because the Mac OS has better secuirty, performance and ease of use (and people get to see that as they run the two systems simultaneoulsy on the same hardware), why would the SW companies not meet the demand? I mean here you've got one computer that in affect turns into a market for both platforms. and if it's true that much of the SW used on home computers is currently pirated PC ware, then it's not like the SW companies are loosing sales to PC owners when these consumers choose to buy the Mac SW version, but rather gaining entirely new revenue from new sales to an expanding Mac market.

MS Monopoly is the keyword.


Windows share is HUGE, Mac OS share is tiny. This is at the root of the problem. If software companies realize that most users run the Windows version of the software on their Intel Mac, because it is less expensive or because they got for free somewhat (work licenses etc.), be sure that the next day it will be no Mac version.

Care to back that up?

(Also, the PPC970 variants, like the POWER4/5, have their nest -- which includes the processor interface -- running at a lower frequency than their core(s).)



There's a lot more to high frequencies than power dissipation -- especially if you're talking about PCBs, not on-chip stuff.

I seem to recall that EFI includes provisions for legacy booting.

Well, so far Adobe and ms plan to develop their software for the Mac platform.

It is hard to see Apple wanting its Mac to just be a booting place for a windos programme. Is there not a vested interest that Apple has in keeping its hardware and software distinct? Surely there is better business sense in Cupertino than to just let things drift to being a windos box.

So it sounds like your thrust is that Apple needs to not just allow windos programmes easily work on Macs, as then cross platform developers will stop producing the Mac version. Such as ms and Adobe. Well, the closure of a Mac version of software is a vital concern, for health of the platform.

The Macintosh division will be out of a job then. Wonder what that gal Roz Ho at ms is thinking about all of this, the Mac division person/lead(if she is still the lead). She thinks that the Mac business unit 'is made up of some of microsoft's most talented thinkers and developers.'

Often the Mac version seems to work better than the windos. I would think that a lot of people would kick about the demise of the Mac version of Office. I don't use it, but many people do.

Actually, he was more subtle than that...

"After Jobs' presentation, Apple Senior Vice President Phil Schiller addressed the issue of running Windows on Macs, saying there are no plans to sell or support Windows on an Intel-based Mac. 'That doesn't preclude someone from running it on a Mac. They probably will,' he said. 'We won't do anything to preclude that.'
However, Schiller said the company does not plan to let people run Mac OS X on other computer makers' hardware. 'We will not allow running Mac OS X on anything other than an Apple Mac.'"

They don't preclude anyone from running Windows on the current Mac, should they want to. It doesn't mean it's easy.

I think being able to use Windows on a Mac is one of the best things they could do.

I am a PC user and a Mac user, but I spend most of my time on my PC because I like Windows and the apps I have on it. I like OS X too (maybe even more), but I don't particularly like using my Powerbook because it's so slow (1.25GHz 15" 512MB etc). I will probably sell it actually to get one of these new Macintels if (and only if) it can run Windows. Gaming is pretty important to me and I've never been a fan of emulation.

The way I see the future panning out - Windows runs fine, lots of people who wouldn't have considered a Mac in the past buy them (like myself at one stage), game development for the Mac dies off a fair bit, user base continues to grow, lots of new apps get made for the Mac, user base grows more, game developers see the untapped native OSX market and the Mac becomes a viable gaming platform. I don't think losing the games in the short term will be that harmful if the overall result is to grow the mac user base.

Well, I've spent over an hour this morning trying to find ANYTHING about the memory controller, but after having gone through over 1,000 articles and multi-page PDF's, I haven't been able to find a single note about the controller.

That's the problem with Goggle, even when you know articles are out there, because you've read them, you can't always find them.

So, if you're constitutionally against believing this, fine. It is true nevertheless, and if I do find what I need, I'll post it. There's really no point in my making it up.

As I've designed hi-freq. devices, I agree with you on your last point. But, it is by no means insurmountable. The main problems other than power dissipation, which shouldn't be a problem for a part designed to run at that speed, is line capacitance, and distance from the chip and main memory. All of those problems have been solved years ago for much higher freq. designs. If Apple wanted to do it they could.

No, it won't have to be easy. Nut, if you looked at Apple's patent application for their new security scheme, it mentions selection from installations of Mac OS X Windows, and Linux.

This looks as though Apple is expecting there to be multiple installations, selectable (upon boot?).

This is from the patent. It, of course, also explains some of the security as well. Scroll down a bit to find it.

http://www.macsimumnews.com/index.ph...esistant_code/

I'm just sceptical because, as I said, to my knowledge the PPC970 variants (and POWER4/5) do not support a 1:1 nest clock, and thus having the EI be able to run at 3GHz would not seem like a very useful feature.

(I'm also not surprised you don't find much info about U3 on Google, given that it's an Apple-proprietary part and all...)



I'm not aware of any digital, parallel, board-level buses that run at multi-GHz clocks on standard multi-layer PCBs...

There's no need to dual boot (which I can't imagine His Steveness would condone) if you've got virtualization technology:

http://news.com.com/Intel+brings+vir...l?tag=nefd.top

Run OS X as your main OS, and run Windows in a separate virtual machine. Thus any bugs/viruses/malware on your windows virtual machine won't affect OS X at all.

Best of both worlds.

Yes, the root of the problem is a tiny market vs a large market,but the way many of you look at this is negative. It's rooted on a series of assumptions
1) that the cost of developing/supporting SW for the Mac platform is higher than the returns
2) that this is not a free market, but rather a monopoly market
3) that minimizing costs is the driving force for making money in the SW biz, rather than expanding income
4) that the Mac SW market is closed, meaning new players-companies or programs-will not come onto the field. (Especially that the expansion of Mac related hardware that inevitably will take place to meet the demand of a gowing Mac platform won't translate into a corresponding expansion in the SW realm)
5) that the computer industry remains static over the next few years, selling/supporting SW in the same manner it has up til now
6) that Apple's emphasis remains on personal computers

Ok, I'm stopping because I realize I don't know enough about computers and economics to make a cogent argument. There is one point I'm trying to make here: that this is a complicated issue and essentially a question of economics and technology. I suspect that Apple has looked at this long and hard, and has had it's own employees as well as outside consultants produce models of what might happen should they go head to head with MS. I can only guess what factors they might put into such an equation. (Actually I can't-anyone with knowledge of high tech economics want to try?)

We are handicapped in that we don't have access to either the big picture, nor the small details. Is Apple going head to head with MS? We don't know. Wlll Apple's prinicple source of income 5 years from now be from sales of personal computers/SW? We don't know. We are also looking at this from a really narrow point of view because this is a very secretive company that in many ways appears to be run more like a privately held family business than a large corporation. What would be the impact on Apple should Steve Jobs suddenly die? We don't know. The one thing we do know is, as a few of you have noted, compared to times in the past where Apple ran into CPU-related hardware crises, Apple's position relative to MS is much improved. Their income and market share are increasing. They are a company on the upswing, whereas MS has been struggling to redefine itself. That has to have an impact on the decisions of SW makers.

I agree with having best of both world, but why does Windows have to load?... I think OS X will run windows apps w/out loading windows.... As long as window's apps run natively/non-natively, this option will lure in pc'ers to macworld.

I can see the new add for Leopard. "Convert to mac and bring your windows apps with you"

Well, parallel buss's have been on the way out for several years now. anyway, it doesn't matter if it's parallel or not, all major bus systems that have been coming in are working in the GHz range. iSCSI, SATA, Express, etc. cell and portable phones both digital and analog, work in the multiple GHz range. One of the scopes I was using has a clock in the 5GHz range. Signal analysers also run at that speed.

The point is that the technology is well understood and implemented. The difficult range is now 10GHz and above.

Maybe you would have better luck than me. Try and find it. Sometimes you can work the search in numerous ways, and only one, just by luck, will find it.This information was out years ago. The controller could have explicitly been designed for dual core chips. IBM had them before the 970 even came out to market.

No!

The ratio CPU to memory controller is 4:1.
What you mean is the ratio CPU to bus data rate which is 1:2 as the bus is DDR.

And bandwidth on the bus is raw transfer rate minus overhead for the packeges. IBM claimed 8:9 for throughput:raw transfer rate.

No, it's not. It's a bi-directional bue. Each direction is 1/4 speed, if you like to think of it that way, but the total speed of the bus is 1/2, or 1/3rd.

The G5 bus is double pumped. Maybe that's what you're thinking about.

All bus's have inefficencies, this one is no different.

Ah, can you say "OS/2"?

IBM made OS/2 run Windows 3.x apps natively under OS/2 2.x in the early-to-mid 1990's. Nobody wrote OS/2 apps, since Windows apps ran just fine under OS/2. It's this very case that is probably prompting many people to warn about running Windows apps too conveniently.

The huge difference here, though, is that IBM never wrote any OS/2 software--not consumer software, in any case.

Apple, on the other hand, makes most of the software that users have on their Mac other than Adobe, Microsoft, & Intuit apps.

Really, other than utilities and shareware/freeware apps, what other software do most people run? (There used to be hundreds of companies producing consumer software; now it's mainly down to Apple, Microsoft, Adobe, and Intuit.) It's mostly Apple software that people run on their Macs, and the percentage of Apple software is growing every year.

In contrast to people who say the Mac experience is the hardware or the hardware/software combination, the Mac experience is clearly becoming a mainly software experience.

No one does software like Apple.

(That's not to say the hardware doesn't count; I just don't think it's quite as important as it used to be. The Mac hardware advantage is mostly style and ease-of-use, rather than technology, IMHO, whereas Apple's advantage in software is style, ease of use, and technology.)

The point about OS/2 is correct. There were other issues as well.

IBM's software division wanted the hardware division to install OS/2 on all IBM machines by default, and offer Windows instead as an option, rather than the other way around, which is the way they were doing it.

The hardawre division balked because they were afraid that most customers, not being familliar with OS/2, would pass on the machines.

So IBM lost their chance.

That was, by the way, when IBM and Apple were collaborating on "Pink" and Taligent.

While Apple does make more software than they ever did, the fact that over 3,000 developers showed up at the conference in June shows that there are many developers.

We really don't want to lose them.

There are far more programs out there than most people think. Most are specialized, but all are important.

Yes I know the difference between data rate and bus clock.

"bi-directional bus"? The bus is dual unidirectional!

One way is 1/4 CPU (=clock of the memory controller) *2 (double pumped bus) *4 (32bit width bus) *8/9 (packeging overhead) = 4.4 GB/s bandwidth in each direction for the 2.5GHz PM.

Hope that makes it clear for you what I mean. Sorry, english was only my third foreign language in school....

That's what we were saying. A bus's speed isn't measured by one direction.

Very good point on OS/2....

The new MacIntel hardware with Leopard being released right around the Vista, where windows users and apps may also need updates to work on Vista.... Everyone will need to go through transitions..... and one on the MS end will predictably be a nightmare.

It will be interesting if Apple can make old window apps run better/stabler on Leopard than the Vista can. At this point in time, all old window apps have to go through emulated process on the Vista as well as if they were to run on Leopard. I'm not even sure if Vista will run old window apps.

In anyrate, it's about gaining market shares on the OS installed base and the developers will follow when that happens.

It depends on what yoummean by "old". Current apps will run.

Apple has the same problem with the Mactels. OS 9 and earlier won't run. It's one way to get Mac users onto OS X for good.

Anyone read digital fortress?

bikertwin and dojobi: yes, virtualization without dual booting means a thin layer that handles virtual x86 machines, without emulation. that would be sweet for windows apps and other legacy rubbish eg. Freecell currently vmware on x86 linux/windows support for 3D acceleration is experimental but eventually developers, maybe with a helping hand from ati/nvidia, will be able to present Directx9.0c & OpenGL2.0 support in virtualization.

this would mean a dearth of mac game development in the short term but i believe as the installed base of x86 mac users grows considerably, new, unimagined-before possibilities will open up after a year or so.

PS. anyone know good freeware Freecell or Solitare for mac os X PPC? i'm serious, its for my aunt that just Switched.

edit: found this freeware. might be good, nice evolution from windoze solitaire to 3D solitaire with freecell, solitaire standard and a lot of other modes!!
http://www.grassgames.com/solitaire/index.html

I like the sound of this virtualisation. I hadn't read much on it to be honest, but if it's as seamless as you say, then I'm all for it.

Can't help you on the freecell front, unfortunately. I understand her addiction, though

Originally posted by dojobi
I like the sound of this virtualisation. I hadn't read much on it to be honest, but if it's as seamless as you say, then I'm all for it.

Can't help you on the freecell front, unfortunately. I understand her addiction, though


heh. yeah. i'm fairly impressed with vmware, compared with just virtual PC in the past. also the Wine open source project has even more potential, as there is no thin client layer at all.

imagine if you could put onto mac os X86 a directx9.0+ library and PC games run seamlessly on it. yes, death of Mac PPC games, but think of the possibilities, everyone....

now think of openGL 2.0 library on mac os X86. now think of game developers going, hmm.. if i write for openGL 2.0 on windows x86, writing for linux x86 and mac x86 is not too much of a stretch... yeah, excellent, f** directX. YEAH.

virtualisation is somewhat in its infancy with regards to games, but for standard apps it is pretty sweet. vmware workstation 5 "multiple snapshot" feature is a million times better and more efficient than the old technique of "ghosting"