Micro$oft to make own CPUs?

[RichL]

Cue the a burst of Orff's O Fortuna* :

Just saw this on the Inquirer (note date, not April 1st!).
Give MS's attitude towards software monopolies, I find the idea of Micro$oft own brand CPUs rather worrying.
I cant see them knocking off Intel or AMD for a while, but who knows?



*Thats the music from Damien, for those of you who dont know.

http://www.theinquirer.net/02040213.htm

Microsoft may have Licence to Kill CPUs

Xbox 2 could further Microsoft monopoly
By Arron Rouse, 02/04/2002 13:11:53 BST


OVER THE LAST YEAR or so, Microsoft has made its intentions clear. It's had enough of selling things and wants to start licensing and renting. The question on all the top executives' lips is, "how can we make everyone who uses or supplies components for a PC pay us a license fee?"
There has been talk of Microsoft producing their own processor for Xbox and graphics chips too (see Eva Glass' original article and the followup here.)

At first glance this seems unlikely. Why would Microsoft risk antagonising the biggest chip suppliers? The answer is licencing. Microsoft doesn't just want to license end users, it wants anybody who produces components for a PC to pay it a licence too.

How likely is Microsoft to produce a graphics chip? The answer is that it is almost a certainty. They have put out a request for quotation for a DirectX9/10 microcode engine. That's the part of a chip that decodes DirectX instructions and runs them. The only reason to do that is if they are building a chip.

Microsoft already has experience of designing graphics chips in the form of Talisman.

You may think that would be the end of it, but no, Microsoft has recently bought the Silicon Graphics patent portfolio. This means it can introduce features into DirectX that require certain patents. All the graphics chip companies that want to keep up will have to buy patent licences from Microsoft. ATI and Nvidia will have to pay if they want to stay in the game.

But what about the processor? Surely it would be far more difficult to produce an entirely new CPU for the Xbox 2? The answer here is that much of Microsoft's plans revolve around the Microsoft Intermediate Language* (MSIL). And Microsoft wants to make Xbox 2 as much of an MSIL machine as possible.

It wants a processor that will run MSIL as fast as possible. The processor will also have to retain compatibility with Xbox 1, so the chip will also have to be capable of running x86 instructions at a respectable speed.

So it needs a processor that can run MSIL like it is native code but can still run x86 code.

Now look at the fact that Transmeta, which is steadily heading into deep financial waters has refused take-over bids . Or rather, it has refused unsolicited take-over bids. Its share price is bargain basement stuff. It's losing money hand over fist. It has a processor that is capable of running x86 code.

The Crusoe can also code morph. Transmeta could easily build a code morphing engine for MSIL - it has already demonstrated a Crusoe acting as an x86 and Java processor at the same time. It starts to make you wonder why two of the top executives at Transmeta left recently. It starts to make you wonder why they were refusing unsolicited take-over bids. So maybe Microsoft has already found the processor they want. And it's a step further into its licensing strategy.

Remember, "how can we make everyone who uses or supplies components for a PC pay us a license fee?"

Don't forget that Microsoft now owns a stack of SGI graphics patents.. and that Nvidia started when some top SGI developers wanted to go it alone themselves.

If Microsoft owns or controls a processor, it can compile Windows to have best compatibility with its own CPUs. It can start introducing patented technology to their processors which Windows "requires." Then it can start licensing the patents to Intel, AMD, et al. Before you know it, MS will be the OS monopoly, graphics monopoly and the processor monopoly. µ

Arron Rouse is a contract technical author and business analyst. Copyright on this article is his.

*MSIL is Microsoft's equivalent to Java's Bytecode. It can be run on any processor provided you have an MSIL Virtual Machine, just like you need a Java Virtual Machine to run Java.

All of Microsoft's .NET strategy revolves around MSIL with Visual Studio (C++, C#, Visual Basic, etc.) capable of producing MSIL packages.


The Inquirer
© 2002 Breakthrough
Publishing Ltd

All rights reserved.

[Jammrock]

Someone alreay posted this in the Crystal Ball or Gen Hardware a while back. Were you actually surprised M$ would attempt this? One of the M$ chairman is a majority holder in Transmetta after all...

Jammrock

[rugger]

Haha, what a joke

Microsoft really doesn't know the first thing about making a CPU.

Take their software as an example :-)

But seriously, what do microsoft have to gain by creating their own CPUs. The market is already REALLY saturated, and developing competitive CPUs needs a real ****load of cash. Especially a good X86 cpu. Transmeta CPUs, as cool as they are, arn't really going to cut it against real hardware implementations (like AMD and Intel CPUs)

In particular, look at how transmeta went themselves, and Cyrix, and Rise, and IDT. Look at the rather pathetic performance of the VIA cpus (trust me, they wouldn't be angling for the power and heat advantages if their chip was really fast). Why do microsoft execs think they could do better, and then charge more for it. If microsoft want to go after the CPU and graphics markets, they will need to realize they will be going against the most experienced and battle hardened companies out there on their own turf (not microsofts). Netscape, compared to Intel, AMD, Nvidia, Maxtrox and ATI was a cakewalk for Microsoft to conquor.

[RichL]

Originally posted by Jammrock
Someone alreay posted this in the Crystal Ball or Gen Hardware a while back. Were you actually surprised M$ would attempt this? One of the M$ chairman is a majority holder in Transmetta after all...

Jammrock

I beat them by 28 minutes, so there.

[Wombat]

rugger, have no doubt that MS could do this if they want. They don't have to do the job well, they just have to get it done, with a project bleeding money like a sieve, until the rest of MS can make life harder for the other players in MS's new market. Within the last couple of years, we've seen the MS Keyboard, Mouse, Joysticks, X-box, MSN....
MS can enter another product line if they want to, since they can be pretty confident that this Presidental administration will let them do whatever they want.

[rugger]

Wombat, its not that simple

Joysticks, keyboards and mice were an easy market for Microsoft to enter. Low barriers to entry, weak competitors with many poor products (look at most mice and keyboards that are on the shelfs today, utter crap, compared to microsoft gear) And still, microsoft doesn't have a stranglehold on these markets.

However with CPU's, microsoft have easily bit off far more than they could chew. Microsoft will have to face very touch competitors who are just as ruthless as they are. Microsoft would not just bleed a little bit from such a venture, but possibly bleed billions of dollars attempting to be an Intel, through the cost of manufacturing and research. Microsoft will not be able to use its other products as leaverage into the general CPU market because:

1) The market has shown that it will not accept any desktop CPU that does not perform well. Saw it with IDT, Cyrix, Rise, Transmeta, and VIA to some extent. So microsoft will need a CPU better than AMD and Intel cpus to even begin to steal market share.
2) If microsoft were to tie its operating system to an expensive subpar CPU, the OS sales would suffer, due to 1. As an example, consider the exceptionally poor sales of Winnt 4 for powerpc and sparc. (mind that there were a lot of factors for that though)
3) People like backwards compatiblity. Microsoft will lock itself out of a large market of upgrade users if it were to release a CPU that only works on MS hardware (I admit the upgrade market is smaller percentage than it used to be too)

Anyway, even in microsoft's playground, software, microsoft hasn't always been able to crush its competition. Take quicken as an example. Due to the complex and constantly changing nature of financial software, quicken has been able to beat microsoft by being more up to date and more complete. As long as Intel and AMD do not stagnate in the CPU buisness, Microsoft will have a very tough time beating them. Thinking about it now, it seems most of the areas microsoft took over were areas when the original players had become stagnate, or forgot to make an important change. eg

MS Dos: IBM was too slow in doing anything with dos, and was too interested in controlling the IBM PC to see what microsoft was doing.

Windows: IBM had not allowed program development to occur under OS/2 as cheaply or effectively as MS did for windows. So MS won by having more/better programs. IBM was still obsessed with controlling the PC market.

Office suites: Microsoft got their windows office suites to windows far sooner than anyone else (the other competitors never beleived windows was serious), and simply took the waiting market.

Browsers: Microsofts browser, IE, became significantly improved upon version 4. It was well tested, fast and easy to use. Netscapes browser on the other hard, was fairly unstable on version 3, and simply got worse in version 4.X. Of course bundling IE in windows was a dirty, illegal trick MS used to kill netscape, but netscape's day was clearly over. And on top of this, Microsoft doesn't make ANY money at all from the browser.

Wombat, you make the assumption that microsoft does not have to do a good job to take over a particular field, however the evidence over the last 20 years suggest not only does microsoft have to do something better than everyone else, they have to rely on the entrenched competitors making a major mistake to open the door for them.

Sure, Microsoft could take over the CPU and graphics markets, but it would possibly require the bleeding of LOTS of money and Intel, AMD and everyone else making a serious mistake.

[spoogenet]

Reading the existing posts, it appears as though people are overlooking another important possibility.

If MS is getting into the processor market....must it be the desktop processor market? Must it just be for the gaming market?

If you look at general trends in the industry, the hot stuff today is mobile, ultra-portable, light, low power, etc etc etc. MS is also working on these markets, take the Tablet PC or whatever it is. MS is flashing that thing around at all the shows these days. They have made huge advances in the PDA (well, Pocket PC really) market in the last few years and will continue that trend. .NET focuses on segmentation and will probably incorporate mobile solutions.

I don't foresee MS getting into the high performance desktop market anytime soon, that's very tough to do. Not to mention how it's bad business practice to enter a market well after its hay day. But it wouldn't be nearly as tough to get into the portable, low-power market which, consequently, will be much more important to their business in the future than the desktop market. Face it, desktops are slowly on the way out as mobile solutions get more powerful and more integrated. I'm not saying desktops are going away completely, but mobile is the hot market for today and the future. Getting into the desktop PC market would pretty much be a waste of money since it doesn't have nearly the future potential or margins of mobile.

b

[Strahd]

I really really hate that company.

[Technoid]

Re: Wombat, its not that simple

Originally posted by rugger

3) People like backwards compatiblity. Microsoft will lock itself out of a large market of upgrade users if it were to release a CPU that only works on MS hardware (I admit the upgrade market is smaller percentage than it used to be too)

That is already shot to hell with XP

Browsers: Microsofts browser, IE, became significantly improved upon version 4. It was well tested, fast and easy to use. Netscapes browser on the other hard, was fairly unstable on version 3, and simply got worse in version 4.X. Of course bundling IE in windows was a dirty, illegal trick MS used to kill netscape, but netscape's day was clearly over. And on top of this, Microsoft doesn't make ANY money at all from the browser.

Funny, when I used NS 3 and 4 my computer wasn't what you call "Rock stable" and NS never crached for me
IE on the other hand was chrashy as hell...

[rugger]

spoogenet, it still doesn't matter

Whatever CPU market microsoft tries to face, they will find entrenced, mean companies already there. In the mobile market, they will find Motorola (with its 68000 and powerpc imbedded processors), StrongARM and ARM processors (does Intel make these). Embedded processors are also fairly price sensitive, since CPU type doesn't really matter most of the time. This market also requires seriously low power devices, devices transmetta never developed.

[spoogenet]

ARM is its own company, Intel makes the StrongARM line of processors. ARM licenses its technology to many many chip companies.

I don't doubt that they will face competition, any company that tries to enter an existing market will face fierce competition. What I'm saying, though, is that I think it's bad business to try to enter into the desktop PC market, and I think MS realizes that.

Despite the competition in the mobile market, I don't think that would stop MS from developing. Just because Transmeta didn't make ultra low power chips doesn't preclude MS from doing it. Transmeta was on the lower power side for their power, IIRC.

Regardless of the competition, though, they can simply take Palm's approach (prior to OS/hardware split). MS already writes software for a variety of mobile devices, and it is looking to get into more. MS is entering into cell phone partnerships, they are touting their Tablet PC, they already write Pocket PC, and who knows what else they are working on at the moment. It wouldn't be a far cry for them to go ahead and develop the underlying hardware platform for their software. If that is the case, there really isn't much competition from other hardware vendors since they will be developing and releasing their own product with their own HW/SW. Who's going to compete with their hardware???

As for cost, if they crank out enough of them using foundries then they won't be too expensive. Low power, well, that's just a technical challenge that a good design team can overcome.

This is all hypothetical, but I think it could be a wise move for MS (ignoring the DoJ, for the moment) to develop hardware for its mobile software solutions and sell the devices themselves. I think it would be unwise to sell desktop PC chips.

b

[RichL]

Re: Wombat, its not that simple

Originally posted by rugger
[B]Joysticks, keyboards and mice were an easy market for Microsoft to enter. Low barriers to entry, weak competitors with many poor products (look at most mice and keyboards that are on the shelfs today, utter crap, compared to microsoft gear) And still, microsoft doesn't have a stranglehold on these markets.

I have to be pedantic and object here. Microsoft mice are pretty damn good. I use an Intellimouse Explorer, and had an MS Wheelmouse before that. But prior to that and at work I've used lots of Logitech mice, and rate them as within an ace of anything MS produces.
Joysticks? Thrustmaster, CH, Saitek, Gravis, do these names ring any bells? I've got a cheapo Boeder stick that does the business.
Keyboards : There are quite a few of us here who swear by the old IBM mechanical (I just acquired one for my pc). Cherry keyboards have long been rated as quality kit.

However with CPU's, microsoft have easily bit off far more than they could chew. Microsoft will have to face very touch competitors who are just as ruthless as they are. Microsoft would not just bleed a little bit from such a venture, but possibly bleed billions of dollars attempting to be an Intel, through the cost of manufacturing and research. Microsoft will not be able to use its other products as leaverage into the general CPU market because:

1) The market has shown that it will not accept any desktop CPU that does not perform well. Saw it with IDT, Cyrix, Rise, Transmeta, and VIA to some extent. So microsoft will need a CPU better than AMD and Intel cpus to even begin to steal market share.

Not neccesarily. The market has been TOLD that it shouldnt accept a slow CPU by some clever marketing. Think about it, do you really, really need a Pentium 4 running at 2ghz to do your daily emails, word processing, a bit of light spreadsheet work, surf the internet, even play a few games of Quake 3?
The answer is no. The only people who really need the sort of power todays processors have are heavy duty users and hardcore gamers. Any anyone who installs Windows XP.

2) If microsoft were to tie its operating system to an expensive subpar CPU, the OS sales would suffer, due to 1. As an example, consider the exceptionally poor sales of Winnt 4 for powerpc and sparc. (mind that there were a lot of factors for that though)
3) People like backwards compatiblity. Microsoft will lock itself out of a large market of upgrade users if it were to release a CPU that only works on MS hardware (I admit the upgrade market is smaller percentage than it used to be too)

Much as I detest Micro$ofts attitude and practices, I dont think it would be that stupid.
It would most likely cut its teeth with a CPU for the X-Box. Then it would release CPUs for the desktop with slogans for the masses like "Bringing the power of the X-Box to your PC!".
All it has to do is ensure its latest programs are optimised to run with a Microsoft CPU. Not incompatible with existing AMD/Intel stuff, just optimised to run better with MS.

Anyway, even in microsoft's playground, software, microsoft hasn't always been able to crush its competition.

Now you really are being silly. At the risk of sounding like Gurm, if you dont use Micro$oft OSes or Micro$oft Office, what the hell else is there? MS have cornered the market and even though there are better products MS's Hype and Marketing department are capable of drowning them out.
Unfortunately, in corporate business (certainly the bits I have to deal with), the IT department has to answer to the users. If the CIO goes to the CEO and says "We want to run Linux version X with Staroffice because...", the CEO, whose knowledge of PCs runs to whatever advert he saw on TV last night says "But everyone else uses Microsoft and MS Office, and all our staff know how to use MS Office..." and telling him about things like common document formats, converters and the fact the users cant possibly be much more incompetent than they already are with another Office suite will fall on deaf ears, no matter how much money you say you can save.

Browsers: Microsofts browser, IE, became significantly improved upon version 4. It was well tested, fast and easy to use. Netscapes browser on the other hard, was fairly unstable on version 3, and simply got worse in version 4.X. Of course bundling IE in windows was a dirty, illegal trick MS used to kill netscape, but netscape's day was clearly over. And on top of this, Microsoft doesn't make ANY money at all from the browser.

I'm sorry, but thats utter crap, IMNSHO. (I really am starting to sound like Gurm )
Netscape 3 was pretty good. IE 3 wasnt.
Netscape 4 was bloody good. IE 4 wasnt.
Netscape 6 was, and still is, a pile of festering crap thats worse than 4. IE 5 actually works and is pretty good.
If I remember rightly, one of the reasons MS actually told the IE coders to do a proper job for once was due to a little thing called Corel Office for Java. Corel (who owned Wordperfect at the time) had coded a version of WP6 to run as a Java application under a suitable browser, like Netscape. This never seemed to get much beyond beta testing stage (I had a version of it) before MS pulled out the big guns and got their asses in gear.

An Office Suite that ran under any Java capable browser had the potential to break Microsofts monopoly in two. A common office suite that no longer relied on you running Windows, or even an x86 compatible CPU.
It would have been great.

[The PIT]

Some good points there Richl.
I know here if you introduce Linux to our place the helpdesk calls would go up ten fold. It just ain't ready for anyone who struggles to press the on/off button of a pc.

[rugger]

Re: Re: Wombat, its not that simple

Originally posted by RichL

I have to be pedantic and object here. Microsoft mice are pretty damn good. I use an Intellimouse Explorer, and had an MS Wheelmouse before that. But prior to that and at work I've used lots of Logitech mice, and rate them as within an ace of anything MS produces.
Joysticks? Thrustmaster, CH, Saitek, Gravis, do these names ring any bells? I've got a cheapo Boeder stick that does the business.
Keyboards : There are quite a few of us here who swear by the old IBM mechanical (I just acquired one for my pc). Cherry keyboards have long been rated as quality kit.

I have to agree with you there, I have a microsoft optical mouse on my machine too, along with an IBM clunker. If you read my post, I said microsoft isn't exactly dominating this market. But they have carved out a respecable niche with excellent products that their software partners could learn a lot from.

Originally posted by RichL

Not neccesarily. The market has been TOLD that it shouldnt accept a slow CPU by some clever marketing. Think about it, do you really, really need a Pentium 4 running at 2ghz to do your daily emails, word processing, a bit of light spreadsheet work, surf the internet, even play a few games of Quake 3?
The answer is no. The only people who really need the sort of power todays processors have are heavy duty users and hardcore gamers. Any anyone who installs Windows XP.

Would you pay more for a slower processor

I agree most people don't need todays processors at the moment. But that doesn't mean there arn't applications around the corner that will need it. It also doesn't mean Microsoft will be able to charge more for a slower processor.

I simply cannot see Microsoft undercutting AMD and Intel enough to make any money at processors, especially at current CPU prices. AMD duron 950's are only $105AU!. Intel celeron 950's are $135. Thats cheaper than they have EVER been, and microsoft wants to enter this market.

I find it much more feasable that microsoft may want to enter the embedded CPU market rather than the desktop CPU market.

Originally posted by RichL

Much as I detest Micro$ofts attitude and practices, I dont think it would be that stupid.
It would most likely cut its teeth with a CPU for the X-Box. Then it would release CPUs for the desktop with slogans for the masses like "Bringing the power of the X-Box to your PC!".
All it has to do is ensure its latest programs are optimised to run with a Microsoft CPU. Not incompatible with existing AMD/Intel stuff, just optimised to run better with MS.

Hahaha, first of all, X-Box is a new system. I doubt microsoft will be introducing a new X-Box for 4 to 5 years (see current lifetime or PS1/N64).
Secondly, unless the new CPU is faster than AMD or Intel, people still won't buy it, and Microsoft probably won't be able to use it in an X-box type system.

Originally posted by RichL


I'm sorry, but thats utter crap, IMNSHO. (I really am starting to sound like Gurm )
Netscape 3 was pretty good. IE 3 wasnt.
Netscape 4 was bloody good. IE 4 wasnt.
Netscape 6 was, and still is, a pile of festering crap thats worse than 4. IE 5 actually works and is pretty good.
If I remember rightly, one of the reasons MS actually told the IE coders to do a proper job for once was due to a little thing called Corel Office for Java. Corel (who owned Wordperfect at the time) had coded a version of WP6 to run as a Java application under a suitable browser, like Netscape. This never seemed to get much beyond beta testing stage (I had a version of it) before MS pulled out the big guns and got their asses in gear.

An Office Suite that ran under any Java capable browser had the potential to break Microsofts monopoly in two. A common office suite that no longer relied on you running Windows, or even an x86 compatible CPU.
It would have been great.

ROTFLMAO You trying to become a comedian.

Netscape 3 was pretty good. It only crashed occasionally and every site worked with it. It had good interactive feel and was fast.

IE 3 was fairly good too. Not as good as Netsacpe 3, but still pretty good. It crashed less than Netscape 3, but was slower and bulkier too. Some sites had problems with it.

Netscape 4 was a disaster. It went from a small, nice browser in version 3, into a massive, bloated monster which crashed and consumes resources at will. Common problems I have had with Netscape 4:

1) Outright crashing on small memory machines
2) Heavy memory usage, long startup time.
3) Often gets into a state where it doesn't render pages anymore, simply displaying a blank screen. You have to close the whole thing down and restart it.
4) Slow as a wet week under complex web pages with nested tables, frames and/or anything other than the most simple of pages.

However, the resposiveness (when it worked) was better than IE 4

IE 4 was really quite good. While it was quite a bit bigger than IE 3, it didn't crash, worked on just about everything (except braindead netscape 4 pages), loaded up quickly (even if you didn't use active desktop), and didn't do weird, unexplainable things. I still use IE 4 on my win98 gaming partition.

IE5 was even better than IE4, but most of the enhancement came in version 4.

Netscape 6 shouldn't exist at all. It was just a stupid idea from the crackpots in AOL (aliens online). Mozilla 0.9.9 is a LOT better (but not perfect)

Please wait again while I ROFL about Corel Office under java again. I can't believe you buy into this java crap. I am sure I tried using a Java based office once, for the whole of 5 minutes that it took me to quit out of it.

Java is a nice language, but it was (and probably still is) far too slow to implement a serious office suite under. Plus, java doesn't magically solve the Document format issues that keep office entrenched.

[RichL]

Would you pay more for a slower processor?

I think you're missing the point.
We, as in most of us here who can be considered reasonably tech savvy simply because we're on a website such as MURC in the first place, probably wouldnt if it was for our own personal machine.
But would I pay a reasonable price for a CPU that does the job rather than one that does far more than I need at a higher cost?
If I had to buy a large quantity for my company, yes I would.
I just put together a 950mhz Duron for my dad. He doesnt need that sort of processing power, but its the cheapest I could get.

More importantly, would a company pay more for a slower processor?
Consider this :
My company, and quite a lot of other big companies, but machines such as Compaq's, IBMs, Dells, etc.
For the price of a Compaq, I can spec a PC using quality parts from reputable suppliers that will blow the doors off said Compaq in speed, flexibility, and probably stability and reliability too.
But these people still all buy their Compaqs etc.
Why?
Because they only tend to buy from the big names, and will buy whatever theyre offered that does the job for the minimum cost.
Is Microsoft a big name? Yes. Will a hypothetical MS CPU do the job? Almost definately. Will some of the big name PC makers build a cheap box with them in? If I know Microsoft, they might not have a choice.

I simply cannot see Microsoft undercutting AMD and Intel enough to make any money at processors, especially at current CPU prices. AMD duron 950's are only $105AU!. Intel celeron 950's are $135. Thats cheaper than they have EVER been, and microsoft wants to enter this market.
I find it much more feasable that microsoft may want to enter the embedded CPU market rather than the desktop CPU market.

Intel and AMD have been cutting each others throats pricewise for the last year or so. CPU prices drop when there is enough demand for them to be mass produced, down to the point where there is no longer the demand for them and production is curtailed.
Did I not say "X-Box" in the previous post? How many X-Boxes do you think MS plan on building, and how cheap can they get that many CPUs produced for?

Hahaha, first of all, X-Box is a new system. I doubt microsoft will be introducing a new X-Box for 4 to 5 years (see current lifetime or PS1/N64).
Secondly, unless the new CPU is faster than AMD or Intel, people still won't buy it, and Microsoft probably won't be able to use it in an X-box type system.

If you look around, there are already rumors circulating about the next generation X-Box hardware being made by X instead of Y, and the damn thing hasnt been out more than a few weeks over here.
Whats the lead time on developing a new CPU? I'm sure one of the Intel guys around here could answer this, but we seem to have gone very quickly from Slot Athlons, to Socket Athlons, to Athlon XPs and now we have the next gen core Athlons looming over the horizon. Similar story with Intel Pentiums too.
All in the space of a couple of years.

Secondly, the current X-Box CPU is an Intel P3/700.
Like I said, it doesnt have to be that fast, it just has to be fast enough and well hyped enough (and cheap enough), and chances are the uninformed masses will buy the thing.
How many people have those keyboards & mice you can get to make your PS2 capable of sending email (or whatever)? Over here a PS2 costs around £300. I put together a Duron950 based PC last week for £308 excluding monitor. See my point?

ROTFLMAO You trying to become a comedian.

Only trying?

Netscape 4 was a disaster. It went from a small, nice browser in version 3, into a massive, bloated monster which crashed and consumes resources at will. Common problems I have had with Netscape 4:

1) Outright crashing on small memory machines
2) Heavy memory usage, long startup time.
3) Often gets into a state where it doesn't render pages anymore, simply displaying a blank screen. You have to close the whole thing down and restart it.
4) Slow as a wet week under complex web pages with nested tables, frames and/or anything other than the most simple of pages.

However, the resposiveness (when it worked) was better than IE 4

I rarely ran into those problems. When I did, I found it was from pages designed to work best with Microsofts own version of HTML and proprietory codes.

Please wait again while I ROFL about Corel Office under java again. I can't believe you buy into this java crap. I am sure I tried using a Java based office once, for the whole of 5 minutes that it took me to quit out of it.

I never said it was good or fast. I said it worked in the beta version I tried, and worked well enough to cause a stir in Redmond. Like I said, it had potential.