🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Thread View: comp.sys.powerpc
2 messages
2 total messages Started by pevans Wed, 13 Sep 1995 00:00
Intel in trouble?
#3639
Author: pevans
Date: Wed, 13 Sep 1995 00:00
63 lines
4527 bytes
Is Intel in trouble?


This might sound rehashed to a lot of people, but I haven't seen any 
debate about it for a while, so I thought I'd take a swing. This is about 
the affect that the IBM/Motorola/Apple developed PowerPC processor 
line might have on Intel's future. Apologies if this sounds dreadfully 
naive.

At present, Intel's flagship is the P5 (Pentium), with the P6 due out 
next year, and the P7 sometime after. Intel have stated that the P7 will 
be the end of the line for the 80X86 line of processors - that, 
realistically, no more computing power can be extracted from that 
particular architecture pathway. However, they have jumped into bed 
with HP to develop a new line of processors, not due out till the end of 
the decade. There was some marketing hype at the time about how 
these processors would transcend RISC technology, but surely HP's PA-
RISC processor capability won't be ignored completely. HP are 
reportedly a little annoyed at Intel's approach to working with them - 
what Intel says goes. I have no idea about compatibility between the 
new HP-Intel processors and the old P series chips.

So far, IBM and Apple are the principle users of PowerPC chips, with 
Motorola and a few Apple clone makers also jumping on. IBM is offering 
AIX, Solaris, OS/2, Windows NT, and almost certainly MacOS 8.0 (to run 
on the new CHRP machines). Microsoft have made some petulant 
comments to the effect that Windows 95 and it's lineal successors will 
never be ported to PowerPC platforms (to keep Intel happy?), but seeing 
as recompiling apps from W95 to WNT is not to difficult, and the 
ability to do so is part of meeting full Windows compliance, I see no 
reason why developers wouldn't offer their software on both OSs. Apple 
are also promising CHRP compliance, so as of next year, assuming their 
promises are good, Apple machines will also run the above mentioned 
bevy of operating systems. Believe it when I see it.

Anyway, the point is that suppose a major PC manufacturer, say 
Compaq or Toshiba or Dell, or whatever, decided that they'd better get 
into the PowerPC game. The PowerPC market could grow enormously 
and do serious damage to Intel. Existing PowerPC processors are 
provably more powerful than the P5 (the 620 is a daemon) and kill the 
P5 stone dead in the low power market (603e, I think). Another factor 
is the 615 chip, due out RSN, which can execute P4 instructions in 
native mode, as well as PowerPC instructions. That chip has a big 
future. Major PC OEMs must be looking long and hard at the PC 
marketplace. Importantly, Microsoft need not suffer at all as Intel 
declines - there's no reason that their marketing emphasis couldn't 
shift from Windows/Intel to Windows/PowerPC, although if CHRP 
machines and the OpenDoc consortium deliver their promise of more 
open and flexible systems, Microsoft's mania for monopolising 
software markets may become harder to maintain. Intel needs 
Microsoft very much more than Microsoft needs Intel.

I'm not predicting anything drastic will happen soon, or that change 
will be swift when (if) it does happen. This is just my own summary of 
the way I see the processor market going, and I'm sure that Intel must 
be worried.


---------------------------------------------------------------------
Peter Evans                                        pevans@rp.csiro.au
Radio Systems Group                             Phone:  +61 2 3724438
CSIRO Division of Radiophysics                  Fax:    +61 2 3724490
---------------------------------------------------------------------
Re: Intel in trouble?
#3637
Author: Robert Cassidy
Date: Mon, 02 Oct 1995 00:00
45 lines
3159 bytes
In article <manders1-2809952304110001@mac35.willets4.swarthmore.edu>,
manders1@swarthmore.edu (Michael Anderson) wrote:

> > "Kind of"?  Let's not mince words.  Open Transport is really one of the 
> > first steps in showing what a new architecture built from the ground up 
> > on PowerPC can do (when it works).  The speed is really nice (and even 
> > then, it still rests on some system code that will be rewritten for 
> > Copland) -- and it will only get better.
> 
> Ummm....as a PowerMac user I am well aware that PPC technology is already
> far ahead of Intel's offerings, and that the gap is only getting wider.  I
> am also aware that Copland will indeed result in a much faster, more
> stable, more memory efficient system.  So it is unquestionable that the
> technological future for Macs looks very bright compared the the
> technological future of PC's and Win95 (although that is not to say that
> their market share future does not look very bright, quite the contrary). 
> However, owning a PowerMac 7200, I am not sure that I would use Open
> Transport as a *premier* example.  It may well have faster throughput, but
> for some reason it takes a really long time to make connections, find
> servers, etc.  And since all the stuff I do is looking up servers and
> making connections through Eudora, Netscape, and Newswatcher, rather than
> large file transfer, it has resulted in significantly slower performance. 
> In fact, I am thinking of switching to MacTCP.  Just my 2 cents. 
> PowerMacs are certainly superior to PC's, but I'm not sure that Open
> Transport itself is that great.

Sounds like a problem. OT should be faster across the board. It certainly
is on my system - Netscape accessing a web page locally but a different
subnet does it almost instantly and it's not even making OT calls. Only
Newswatcher 2.0 and Anarchie 1.6 use OT AFAIK. Eudora has some serious
problems with OT but Netscape does alright. 

With Anarchie and OT, I commonly get 200K/sec or better downloads from
amug and local servers - MacTCP had a hard time hitting 100K. Establishing
connections is extremely fast. Are you using PPP or SLIP in there? They're
still pretty flaky.

OT is that great. It will become evident when they shake out some of the
bugs (it still has way too many) and when it is really put to use. The
concept of networking without the software having to know what connection
is in place should make the Mac a pleasure to use.

-- 
Bob Cassidy
UC Irvine
Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads