Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
AidenShaw said:
The problem would be that there would be no applications available. It's hard to market an operating system if there's nothing that runs on it.

That's entirely the point of .NET - they didn't copy Java for nothing ...
 
oops, I missed something

X-Baz said:
That's entirely the point of .NET - they didn't copy Java for nothing ...

Have I been wrong in believing that .NET was high-performance native compiled code, and not pig-slow bloated interpreted byte-code?
 
AidenShaw said:
Have I been wrong in believing that .NET was high-performance native compiled code, and not pig-slow bloated interpreted byte-code?
.NET is really neat that way; It has really 3 ways to run: native compile, a really nice JIT compiled (default), and an 'interpreted' mode.
Reason for that being is you can write C# code, VB code, and managed C++ code and tie them all into one program; so you need some 'mid level' interpreter for all 3 so they just opened that up instead of making it an internal feature of the compiler.
 
AidenShaw said:
Have I been wrong in believing that .NET was high-performance native compiled code, and not pig-slow bloated interpreted byte-code?
it's a bit of both - it is an implementation detail as to how it is run - interpreted, JIT-compiled or compile on install.
 
Why IBM doesn't make x86 processors?

Afaik, the original agreement between Intel and IBM, stating the presence of Intel chip inside of the IBM PC, said amongst other things that IBM can't sell their own processor with x86-compatible microcode. IBM tried once even going around it via offering processor cards with their own 486 clone (or was it 386?). Do I remember the things right? Anyone else recalling similar stuff? This may explain pretty much, why IBM is pushing PPC so heavily.
 
IBM need not go "head to head" with Intel in the x86 arena specific to the consumer market (that is, the Pentium/Centrino series). What they can go after is the high-end Xeon servers, which they are doing now with the POWER Series. Even Sun has learned not to go against Intel in the low-end server division, where they now are selling Intel-based solutions, delegating the SPARC-series to the high-end market (again, competing with Xeon and POWER). They can also go head-to-head with Intel in the embedded chips arena, where Motorola is also trying to regain (?) its foothold. Intel has always been the giant in that market, powering up millions of devices released each year, from cellphones, to teleivions, etc.

Actually, it wouldn't make too much sense for IBM to even produce an x86-compatible chip much like AMD does. They have Apple now for the "consumer market" (in terms of competing with Intel at that level), and I doubt Microsoft would fancy yet-another-chip to support (since it won't entirely allow them to crush the competition).

But in all this, its IBM who ends up the winner. At the end of the day, what matters most to them anyway is products/services being sold. So long as their customers are happy with their products, they're happy (since its money in).
 
Healthy Competition !!

Even though the PPC and x86 are significantly different architectures, they both provide the same end result - they power our computers. The tide always turns and aside from Motorola perhaps failing in its chip manufacturing devision, the PPC is on the crest of a wave and IBM and Apple both had the vision to pursue this relatively young architecture. x86 is old and perhaps AMD and Intel are hitting some development barriers. It is good to see so much interest in PPC in other mainstream products.

Just as makers like AMD are important in keeping competition alive in the x86 arena, IBM is now offerring real competition to x86 with some fantastic PPC chips. This is all good for the consumer - more speed and hopefully more competitive prices. Healthy competition I say !
 
Macrumors said:
The fact that Apple had considered using the Intel processors for their PowerMac line was previously revealed in another internal IBM publication.
Thank you IBM for stepping up to the plate.

Back when the rumors were going around, I had faith that Steve Jobs, like the rest of us, had enough pride left in him to prevent Apple from becoming the whipping boy of the personal computer world.

We may have been knocked down for a while, but I'm glad we found the energy to keep fighting.
 
jouster said:
Ahh..because LotD has been such a roaring success?

If LotD means "Linux on the Desktop" then I would say yes, very much so. It has most all of the same type of office programs that the vast majority of corporate staff or people in education and other sectors would use. Plus it is supported by a very large, dedicated base, is mostly hardware independant and quite cheap to acquire. (I don't know about the running costs). The only reason I'm not using Linux (as a backup) is because Macromedia and Adobe havn't ported their apps. Not that I dislike MacOSX :)
 
Did any one else click on the picture of the processor card left of the title?
There's two different picture that come up (do a refresh). One with some fingers in them which sends you to an Intel page and one with out which sends you to a flash animated page of bunnies doing a 30 sec re-enactement of the Exorcist.

I found that kind of strange.
Amusing though.
 
I feel like the nerd boy at the homecoming dance being fought over by the head cheerleader and the prom queen.

Good times...good times.
 
AidenShaw said:
...
You can't find many of the important commercial apps running on PPC Linux, however.

Well, given that there is little more difficulty than a simple recompile involved in going from x86 Linux to PPC Linux, I think that if there were some major manufacturers making PPC Linux boxes, I think that many of the 'important commercial' Linux apps would be released for them.

(Similar to the fact that a lot of the 'important commercial' Linux apps, as well as many scientific apps (which are, effectively, free), are being recompiled - with some minor modifications - to run under Mac OS X.)
 
1macker1 said:
I don't think Intel is really bothered by this, it's just IBM. I think they are more concerned with AMD.

IBM is a heck of a lot bigger threat than AMD is. AMD has one fab and two high performance chip architectures. I have no idea how many of each IBM has, but I can think of three high end chip architectures off the top of my head: POWER4, POWER5, and 970 (750GX doesn't really count as high performance at this point). AMD is doing pretty well right now (beating or matching both Intel and IBM), but they're still a lot smaller than IBM, as well as having a less advanced manufacturing process.
 
Catfish_Man said:
IBM is a heck of a lot bigger threat than AMD is. AMD has one fab and two high performance chip architectures. I have no idea how many of each IBM has, but I can think of three high end chip architectures off the top of my head: POWER4, POWER5, and 970 (750GX doesn't really count as high performance at this point). AMD is doing pretty well right now (beating or matching both Intel and IBM), but they're still a lot smaller than IBM, as well as having a less advanced manufacturing process.

IBM could be a threat, if they wanted to be. However, the internal strategy changes far too often and they will most likely move back into the shadows, rather than compete one-on-one with Intel. AMD designs with IBM processes will provide a better threat.
 
AidenShaw said:
Have I been wrong in believing that .NET was high-performance native compiled code, and not pig-slow bloated interpreted byte-code?

JIT compiled byte-code isn't slow, btw -- specially for "Enterprise" Type applications. In this day and age, the only benefit to machine-speicifc compiled code is if you're doing serious number crunching and need the CPU specific instructions for pure computational efficiency. Distributed computing (a-la .NET, J2EE, TPMs in general) don't need it. Plus, if I can have far better hardware independence. (Dev on Intel, Deploy on Intel, Sun or IBM)

Pig slow applications in Java are "designed" that way... Good software runs just perfectly fine fully interpreted -- even 7 year old Smalltalk applications!!
 
The "Cell" CPU

PlaceofDis said:
i just wonder where/how IBM is going to expand next? they are an interesting company but i have never really kept tabs on them too much, i wonder if they are going to try to go head to head with intel?
---------------------
Havent you guys around here heard about the promising "Cell" processor
which will be out in 2005 ? Is being developed by IBM, Sony & Toshiba !!!
Will be in Playstation 3, and coming computers. 1 Teraflop ANYONE ???
IBM is now looking for someone to develop an OS for the coming systems.

I beginning to love Apple and am going to purchase a Mac G5 somtime.
But news as the "Cell" processor being out in 2005 leaves me a little
annoyed.. since well.. not all is bout processing power but i really need
it since im drooling complex 3D scenes in 3D applications such as Cinema 4D.

:confused:
 
spinko said:
If LotD means "Linux on the Desktop" then I would say yes, very much so. It has most all of the same type of office programs that the vast majority of corporate staff or people in education and other sectors would use. Plus it is supported by a very large, dedicated base, is mostly hardware independant and quite cheap to acquire. (I don't know about the running costs). The only reason I'm not using Linux (as a backup) is because Macromedia and Adobe havn't ported their apps. Not that I dislike MacOSX :)

Don't get me wrong, I don't think suggesting Linux as such an OS is in any way a bad idea. That said, and despite all the sensible points you make re corporate computing, the Linux-on-the-desktop revolution always seems to be real soon now.

If Linux is supported by a 'very large, dedicated base' then Windows is supported by a Gigantic, dedicated (in the sense that people are unaware of any alternative) base.

What will change that, however good a new processor turns out to be?
 
3Dfx_man said:
---------------------
Havent you guys around here heard about the promising "Cell" processor
which will be out in 2005 ? Is being developed by IBM, Sony & Toshiba !!!
Will be in Playstation 3, and coming computers. 1 Teraflop ANYONE ???
IBM is now looking for someone to develop an OS for the coming systems.

I beginning to love Apple and am going to purchase a Mac G5 somtime.
But news as the "Cell" processor being out in 2005 leaves me a little
annoyed.. since well.. not all is bout processing power but i really need
it since im drooling complex 3D scenes in 3D applications such as Cinema 4D.

:confused:

I'm sure everyone has heard of it, but there are lots of unkowns that hamper serious debate:

1. Most rumors (fwiw) have it arriving at least a year late. This would seem to be reasonable - when was the last time such a huge project was completed early? The fact that it might have to wait for the 65nM process to get acceptable yields for its high end one billion transistor (!) procs would seem to support this. It is one of the main objections that can be found at a number of sites: if the most powerful (16 core as far as we know) version has 1 bn transistors, how can it be produced cheaply enough for a console? Or for a personal computer? Or, if the intended processor is scaled down (say, 4 or 8 cores), will it really be a big win over whatever speed non cell architecture will by then have reached?

2. There is now a lot of confusion surrounding its use in the PS. Some say PS3, some say PS4.

3. There is a lot of skepticism over the 1tflop claims - just go to ars for example. That's understandable too; it's one thing to stretch Moore's law, but quite another to shatter it - even with such a new architecture.

4. There are no confirmed customers, just rather vague statements concerning scalable use in anything from PDAs to clusters.

5. As you point out, there is no mention of anyone developing an OS, though one would assume that the PPC architecture would make it viable for Apple.

6. What about altivec? Would one of the cores be a vector processor? Would the multi-core architecture rule out the need for it?

7. How would the OS need to be altered to take advantage of the parallelization? Would it be able to do so without a major rewrite? Remember, being multi-processor aware is not the same as being multi-core aware. One of the main problems foreseen for the Cell based PS is the complexity of the planned software. PS2 software was apparently a nightmare to write; most tech sites think that having decent SDKs and other tools would be absolutely essential for Cell.

The best bit of information we have is simply that it really exists. Even the G5 remained a rumor (GPUL anyone?) for such a long time. I think its an exciting prospect, and with so much money and expertise being pumped into the project, I'm sure it will succeed. But I would bet that: (a) it won't achieve these outlandish specs until at least its second gen, and (b) it'll take a lot more than impressive specs to make a PPC platform outperform whatever Intel has by then - at least in the marketplace.

Maybe its scalability is the answer. We'll see. Like most here, I would love to see a proc like this in a Mac - anything to counter this insane power/heat/Mhz curve.
 
its very funny how everyone dismisses Intel or AMD. Its still true that Macs are being hammered by the otherside, G4 i wont even go there why bother. G5 we still compare 2 of those vs 1 of the otherside and the otherside can still match those 2 G5s in most aspects and even beat them in others. I noticed know one is mentioning that Intel has chips using light in prototypes while we are pushing electrons. Microsoft is like a big bully and when someone else gets big they dont know what to do. Intel has gotton so big that microsoft cant dictate to them. Intel is still #1 and it didnt come by selling slow cpu's such Apple has pushed on us for years. It came by selling fast cpu's and by having more and more of the market wanting their product. seen any G5 is the fastest PC commercials lately? I didnt think so.
 
Dont Hurt Me said:
...G5 we still compare 2 of those vs 1 of the otherside and the otherside can still match those 2 G5s in most aspects and even beat them in others...

Gee, if you look at the comparisons on this Apple page, it certainly seems that they are comparing singles to singles and duals to duals, such as

charthmmer_11172003.jpg


Or

chartbibble_11172003.jpg


... seen any G5 is the fastest PC commercials lately?...

As a matter of fact, yes. I saw one about a week and a half ago.

...Go troll somewhere else...
 
Snowy_River said:
Gee, if you look at the comparisons on this Apple page, it certainly seems that they are comparing singles to singles and duals to duals, such as

charthmmer_11172003.jpg


Or

chartbibble_11172003.jpg




As a matter of fact, yes. I saw one about a week and a half ago.

...Go troll somewhere else...
And now we only have to wait till someone comes in and says: But Apple crippled the PC's scores, to make them look bad and the G5 look good. :rolleyes: But it won't be me :p
 
if you want to pull benches dont get them from Apple. they spin more then my quicksilvers fans. go to some real world places such as Mac World magazine, Mac Addict Magazine and ARNS or Toms Hardware. Apple stretches a lot of stuff and has a very bad habit of finding those benches that favors their product. Remember the fastest computer adds? Ill go look for a few benches.
 
strange

iggyb said:
I was watching some IBM commercial the other night, and they were talking about their Blade Servers running Intel Xeon processors. This made me wonder, why aren't they offering servers with their own processors? Does it have to do with the software involved :confused: ?


I dont know, but i noticed that too
 
I decided to not bother with benches, you can pull benches to show pc is faster and pull them to show 2 G5s are faster. what it boils down to is we still use 2 cpu's to compare with 1 from the otherside. that alone should tell you something and i have yet to see real world benches where single G5s can match single cpus the otherside has. Intel is number 1 and there are many reasons for it. Apple isnt running those fastest PC adds and there are reasons for that.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.