Slashdot Mirror
National Semiconductor Selling Cyrix
Rob Miller writes
"National Semiconductor has announced the sale of Cyrix. "
So if you have half a billion dollars burning a hole in your
pocket...
← Back to Stories (view on slashdot.org)
← Back to Stories (view on slashdot.org)
This sucks. We had three viable chip manufacturers out there, and now we're down to 2. Not good. Cyrix made good, cheap chips that are completely useable, and cost about half of what Intel stuff costs. This sucks.
Via eBay, I suppose. :-)
Yes, Bombardier is a Canadian company. Only in Canada do successful aerospace companies recieve huge government tax credits.
Time to trade in the Learjet. :-)
i got a Cyrix MII-333 for my girlfriend for $109 AUSTARLIAN, which is ~$65 USD equivelent, you cant say that is not better than AMD. Its just a tad 2% faster than my overclocked P233MMX box clocked to 266 (75*3.5)
.au
AMD is not an option since its 2.2 volt only and my older ABIT boards dont go below 2.8 or so.
Cyrix may have crap FP, but for general use its damn good for $109, compare that with $350+ for a PII-300 here in
MS could tell Intel to kiss off and start selling Microsoft sealed boxed PCs with a win2k bootstrap in bois, the rest on a 4 gig hard drive with office and msn internet service. Sell it at Best Buy, Sears, K-Mart, and Wal-Mart (USA) for $399 and everyone who can't program a vcr can own a computer. Personally, a comet hitting the earth would be nicer.
I am sad to see Cyrix go. I have had only Cyrix chips in all my computers even with my 486. I love Cyrix chips because all the extra money I save goes toward kick ass graphics cards, and lots of memory. Long Live Cyrix!
Cyrix has been dying for a long time. The just let too many people down, too many times. I owned one of their 486 DLC/33's, and one of their 486 DX4/100's. Those, and the half a dozen other Cyrix based systems I had the misfortune to work on all had no end of problems. I hear the 6x86ML, or whatever it was called, finally solved the heat problems they had been plauged with, but it was already too late in my book.
For my two previous upgrades for my PC were Cyrix/IBM chips (a P200+ and MII 300). I work for a computer shop and we found these chips to be excellent (we used them as workstations in small networks). Now we have to go to AMD (a gamers chip) or Celeron (too expensive to be a cheap viable solution).
I think the problem with the Cyrix chip was its FPU. That was a well documented thing about the Cyrix FPU and its weakness compared to the Intel Pentium. If Cyrix increased the power of its FPU, and make hardware reviewers see the difference, then they would have survived. When was the last time Anandtech or Tom's Hardware page used a Cyrix in their reviews.....
Lest we forget.
It's too bad you had problems with your Cyrix systems. All my evidence has shown that these were always either motherboard BIOS/chipset incompatibilities, or the CPU not recieving enough power. To run a Cyrix chip, you really need a good mobo... When I bought my second "pentium" system, I specifically told the retailer I was going to set up a 6x86 system, and would like parts that were fully compatible. Needless to say, I never had any more troubles with that system than my P60. My MediaGX laptop works flawlessly (albeit, slowly); and my 6x86 150 Server just rocks (what do you need FPU speed in a server for?)...
:-( Many people thought it sucked... But it was a full 486 compatible chip, and just needed a bit better hardware.
;-) Think McLaren F1 engine in a Beetle......
The cyrix chip was very compatible, but very hardware demanding...
Infact, the 6x86 cpu did show a little more feature than early Intel Pentiums. It has a 0 cycle wake up from sleep mode cost. This helps the laptop tremendously (50% battery life increases!).
As far as DLCs go, what sort of compatibility/performance can you expect from a 386 architecture with a 486 plopped into it...
I can't comment on the Cyrix 486 DX 100 though. I'll take your word for it that it was slow...
Buying Cyrix could make sense for IBM:
1. Buying Cyrix would give it its own rights to an x86 architecture. My understanding is that IBM does not own these rights, but pays royalties to make x86 stuff.
2. IBM will soon have $5 Billion in spending money after the sale of the IBM Global Network gets through all the red tape.
3. IBM had been fabriacting Cyrix chips for several years, so there might be not retooling at all.
4. IBM chip business is among the fastest growing in the industry: the addition of an extra Fab would be a boost to its production capacity.
5. If IBM owned Cyrix, the Cyrix chips in its computers could squeak more profit in a busines which seems to be giving Big Blue a bit of trouble (losses approaching $1B FY1998).
6. x86 chips would be a strong offering in its OEM business. IBM could use the Cyrix platform to increase its share of the OEM market.
7. PLUS, more patents and intellectual capital
Sound good? I would fully expect to hear IBM looking into this purchase...
AMD needs to jump on this. This is just the break that they need. With cyrix under their belt they may be able to increase their production capacity to finally give intel a true run for their money. Anyway, who is going to miss cyrix?
NS/Cyrix can't honestly expect IBM will continue to make Socket7 CPU, can they? If the deal doesn't include Slot 1 key technologies that National has, whoever buying Cyrix are getting screwed.
Unfortunately you could never build a system at those prices.
...
Those may be the best prices listed on Pricewatch but if you shop on Pricewatch you will find that the best prices listed there are complete BS 99% of the time.
Sure, you may find someone willing to sell you a chip for $31, but only if you'll agree to have it shipped for $15. Unscrupulous vendors listed on Pricewatch (which is most of them) will hide the true price of the product in the shipping charges.
I called one place to ask how much it would be to ship a MB. $20. Yikes. Ok, well how about if I add some memory to make it more worth the while. What? An extra $9 shipping just to put the memory sticks in the MB box? Forget it!
I would add 30% to all of the figures you listed to get a more realistic price for the components shipped to you
In case you can't tell, after being a big Pricewatch fan up to about a year ago, I think Pricewatch is almost completely worthless now.
Umm... hate to tell ya, but I have a local vendor that sells stuff close to those prices. It is possible to build a system like that for under $300 for most anyone in this area (SLC, UT).
Didn't Sun make an announcement about this a couple of months back. I think it involves the same "Community Source" approach as Java. Meaning you can look at the source, but it you want to fab it, you have to negotiate royalties with them.
> That's because Linux is so fast...
Linux isn't that fast, is the disk subsystem that is slow. Linux don't use the CPU that much. Try running top. 98% free CPU time.
Need to say anything?
550 people were sacked. Cyrix in terms of the MII product and other socket 7/370 parts will be sold, so have not been sacked. 550 people all over national have been sacked. Full conference call here: http://209.67.65.42/ccall/nsm/050599.ram
This is a shame, though I can't blame them.
I've build a variety of Linux boxes on the MII, simple because a lot of people I know needed something cheap that worked halfway decently.
I could put together a workable system for a couple hundred dollars.
MII-333 $ 31.00
TX-PRO II MB 51.00
32MB SDRAM 32.00
3.2GB HD 79.00
36X CD-ROM 32.00
3.5" Floppy 15.00
AT Minitower 19.00
Around $250 excluding monitor, keyboard, and mouse. And I would normally pick up an honest to god IBM keyboard for four or five bucks at the computer shows, a cheap five dollar mouse, and let them decide whether to buy a decent monitor or pick up a used 15" at the computer shows for around $40 a piece.
Well, in the x86 chip market, it seems we're down to three now - Intel (still the leader), AMD (catching up), and WinChip (whose chips, despite the name, work in OSs other than Windows).
I personally have a Pentium II 266 and a Winchip C6 200, but I'll probably get a K7-based system when the K7 comes out, so I'll have one from each. Competition is good.
PS - why is the WinChip 200 so damn slow? I don't use that computer a lot, so it doesn't really bother me, but it seems to be about the same speed as my dad's pentium 120 laptop...is it supposed to be that speed? Any idea what the Pentium Rating would be on that? (I like how Cyrix does that...makes comparison a lot easier)
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
I don't think Cyrix is going to survive in the x86 market. Everyone who is trying to compete with Intel is doing so by losing money. The only company making money in the x86 market is Intel. Companies like Cyrix cannot survive in this environment.
I remember hearing a year or two ago about Cyrix's "Jalepeno" core, but after a while I came to realize that it just isn't going to happen. Cyrix is too far behind now to catch up. I've been wondering what they have been doing for the past 18 months when they have introduced no new designs.
I put a Cyrix PR200 into a computer I built for my sister almost two years ago. It is still going strong and still meeting all of her family's modest needs. At the time, it was far and away the best value in the x86 market. Now I think AMD has surpassed Cyrix in bang:buck ratio. You can get a K6-2 350 for about the same price as a MII-333, and the AMD chip will beat the Cyrix chip in every way.
This didn't stop me from putting a Cyrix PR300 (overclocked to PR333, although I suspect that Cyrix is selling the exact same chip as 300 and 333 just to try to differentiate the "top end" of their product line a little bit) into a computer that I built for my next-door neighbor for use at his business. That level of performance will be more than enough for the simple real estate management package he needs to run, and combined with a super-cheap built-in audio & video socket 7 motherboard, I was able to keep the price of the computer very, very low (sub-$450).
I guess that if Cyrix goes away I will miss them, because their chips have always done very well for me. I just don't think that they're keeping up with the market at all. And they're at the lowest of the low end, where it's nearly impossible to make a decent profit.
Definitely sacked.
The press release is here.
(So hah, you're wrong, Jordy)
Carl
well, the timing is a bit off - especially with the $5B payoff-er investment into AT&T to corner the settop box market with CE.
Otherwise, if MS somehow slimes their way out of this DOJ thing (there's GOT to be an angle), I could easily see an endgame scenario of MS buying the remains of Cyrix and that would be all she wrote.
Perhaps an intermediate holding interest would arise. Someone stupid like Compaq could buy them, then after two years of hemorraging, while MS shakes off the DOJ, MS picks up the remains for a song, and. . .
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
Personally, if I had large sums of money to spend, I'd be much more interested in that De Haviland Mosquito over in Canada that's for sale. No more parking problems!
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
From the National posting, it sounds like National is selling a fab and giving up on x86 pin-compatible architectures (ie: MII and Jalapeno). They will continue to develop the MediaGX line of systems. Further, they will refocus on information appliancs (like that web pad thing that Cyrix is developing).
--Lenny
//"You can't prove anything about a program written in C or FORTRAN.
It's really just Peek and Poke with some syntactic sugar."
Yeah, my father works there. NS employs a lot of people in Maine, and they could all lose their jobs because the faceless, brainless execs on the west coast wanted to play empire and try to compete in areas in which they were not competent to compete.
-Eric
Maybe we should define cheap? I cannot help but think how much each one of those transisters costs. Last time I checked, Electronics 101 used the transister as the building block for circuits.
;-).
Now we have a company like Cyrix using Memory Controller, Video Card, PCI Bus Controller, Display Controller, etc...
If Moore's Law continues for the foreseeable future, where will it lead us? Are you seeing the trees or the forest? I think openning the data sheets will allow the gifted technologists to design and develope systems that create entire new industries. When the first copy machine was sold to business, did anyone envision how many people would use it to distribute Grandma's brownie recipe?
Saying that Cyrix mainly integrated 3D graphics, or Open Source has a negative effect by eliminating jobs, is similar to judging a dangerous iceberg by the portion you see above the water
since amd has been complaining about only one fab (i guess fab 30 isn't finished yet), here's a great opportunity to get a new one, and keep up with the demand...
i could be way off base, but i don't think it would make that big of a difference, because who is going to donate a fab?
cyrix was always the ugly duckling of the cpu world. They should stick to real innovations like the webpad instead of chips with phony mhz ratings. We need competition to intel and AMD is the only one with a chance. It needs to be stronger.
---
Linus, tell your boss there is a fab lab that has .18 micron capabilites going for cheap in Maine. Time for some prototyping or sample runs? ;)
It is far enough from Redmond and Santa Claira I think.....
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
Good idea. Integer performance has outpaced the software on all but the bloated Microsoft OS's. Since Linux and Internet Appliances are the future it makes sense to use more transistors for FP performance boosts and leave the integer unit cranking at arount 200MHz with some fast onboard cache. I've been telling everyone to buy the PC around the 300MHz range and spend the extra $$ on the graphics and disk IO system (SCSI). Always good to hear differing ideas, keep em coming.
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
National Semiconductor is getting spanked by Intel. NS makes a lot of different products which could make a lot of money. Competing with Intel is expensive.
If tits were wings it'd be flying around.
Back in the early 1990's Cyrix had superior floating point performance.
Too bad they didn't keep that advantage. Floating point performance is more important than CPU performance for many games, and it would have made a difference in my processor decision.
I bought an Intel Celery 300a because of the floating point performance. I will probably buy a K7 in the future because it looks like the K7 will beat Intel (but who really knows for sure?)
Running Linux, my 300a Celeron isn't any faster than my Pentium 133. The system feels exactly the same. Compiles go quicker, but X runs just as fast on slow chips. That's because Linux is so fast on just about any computer, it's hard to tell the difference from ludicrous speed and "plaid." So, for my purposes, I don't care a bit about how fast the CPU integer performance is. The extra floating point kick makes a big difference when I am gaming though.
Cyrix could have carved out the entire gaming niche for itself if only they had built a CPU with P-200 integer performance, but with super duper floating point performance.
If tits were wings it'd be flying around.
Now wouldn't that kick some hind-ends.
--bricktoad
My friends, we are nothing but wings on the chicken of society.
NS' bean counters didn't expect there to be as much competition in the sub-$1k market as there turned out to be, so they bailed. The local news story is here. That link will change tomorrow (Friday EST), but it includes a picture of the South Portland plant.
For chip historians (chip wonks?), the original Fairchild plant is to the right of the building as seen in the photo, just outside the frame. It is currently home to the new Fairchild spin-off.
Save Maine's economy: write stuff down. All comments are exclusively my own, not my employer.
to me, the most telling thing is that NS's stock _rose_ a buck fifty on the announcement that they're divesting themselves of that nightmare processor division.
i browse at -1 because they're funnier than you are.
Now let me think, who might be in the market for a chip fab? TransMeta comes to mind... ;)
Of course, I'm totally ignorant about chip fabrication... Would they need to retool?
-- What you do today will cost you a day of your life.
Just to comment- I have a winchip 200mhz running in a cheapo motherboard (430tx?), and its performance under linux is very snappy. I've got 64 mb of ram and run KDE and am constantly impressed with how reliable and fast it is.
Nothing new. Back in Singapore, NS has started making the Cyrix's since at least 1.5 years back. Now, they've just concluded their 1st round of retrencement, with a 2nd to follow.
A pal of mine quit the engineering line because he felt the management wasn't going anywhere ever since they started making the chips, and since then they started losing $$$ (my mum's still there).
I personally would like Cyrix to be a 3rd chip available in years to come (along with Intel and AMD), and though such announcements hardly call for attention, here's wishing them all the best as well with their new (if at all) strategies.
--9102--
Well, I'm not so sure about that. Most people think that the whole semiconductor industry started out in California or Texas, but the truth be told, it all started in Portland out of Fairchild Semiconductor. So essentially, they were not diving into something new, rather just getting back into something old.
The worst part of the whole situation is what it's going to do to the South Portland, ME location. The news.com article also mentions that they are going to be selling a controlling interest in one of their fab processes too...
This sucks for my friends who are supposed to be working at National this summer. Doh!
It would be cool if IDT baught them out.. Bet the would have some idea what to do!
"Windows 98 Second Edition works and players better than ever." -Microsoft's Home page on Win98SE.
I ate my tag line.
-=Ellis (D)25=-
IDT....
Their processors are quite funky.. I own a 240, non mmx.. It acks wierd at times.. Doesn't know if it's a pentium, 486, cyrix, 6x86, 5x86... that's what comes back on some detection programs.
"Windows 98 Second Edition works and players better than ever." -Microsoft's Home page on Win98SE.
I ate my tag line.
-=Ellis (D)25=-
After playing w/ real IBM PC's, not the clones.. I find that they should follow the lead of other computer companies, but they do wierd shit like ast.. I personally don't know any one w/ a real IBM pc. Just clones.. IBM owning this would only IMO would just screw it over..
"Windows 98 Second Edition works and players better than ever." -Microsoft's Home page on Win98SE.
I ate my tag line.
-=Ellis (D)25=-
Somehow, I think that's probably a bad idea. Open Source works well because compilers are cheap (free for most of us) and its easy for just about anyone to write new software. Also, people can still make money adding additional features to Open Source projects for companies that are willing to pay for it.
But, at least until chip fabrication is cheap enough for a small business to afford, the only thing "open source" chips would do is eliminate the jobs of lots of computer engineers or cut their salaries in half... and you know what that means... more engineers going back to school for their MBAs!
This may be true if they bought Intel (or even AMD if they continue growing) because they have the masses attetion. But with Cyrix, no one knows who they are so why would they buy it? And also if they bought it they would probably screw it up (like most of there software).
Beau C
I bought a complete Cyrix P150+ system w/32MB of memory for $50. It upgraded my AMD 5x85 133MHz server (mind you, I'm running a tiny network in my bedroom, and basically use the server as an FTP archive and dialup IP masq host). It is one heck of a machine, although it does run a bit on the warm side, but nothing too bad. It keeps up with requests from my PIII-500, P2-450, P2-233 and the 5x86-133 just fine (although I wish I had a cable modem). They might have been better off if they found a way to run their processors cooler and with a faster FPU. A little bit of marketing might have helped too! Sad to see them leave....
..Amd has almost no money left.
I just hope they can get out the K7 and keep
up with any initial problems, because we
seriously NEED a competitor for Intel, and
AMD is the only x86 chip maker with enough
innovation.
I _know_ the K7 will be a killer chip..
..some people try to argue that "why should AMD
be able to do it, when Intel isn't"
The answer is simple I think:
Almost all R&D at Intel is going into the Merced, and not much is left for creating a 7th generation
x86 cpu.
The result of this for AMD is pretty interesting, as they are now the major competitor for Intel (not that they haven't been for the past year or so). Even though it is conceivable that Cyrix could be bought and rematerialize as a strong company, as another poster pointed out, that's extremely unlikely.
Research takes a *lot* of money, and because of this news that Cyrix is being sold, it seems extremely obvious that they are short on cash (and probably ideas).
So has Intel won? It's interesting to think that Intel has established such a strong brand and product line that they've done what many predicted they wouldn't be able to do: defeat their competitors.
Hopefully this'll wake some people up over there at AMD, that all the patting on the head they've been giving themselves due to their moderate success doesn't mean much-- they haven't really beaten Intel, but rather just established a niche for themselves. But Intel isn't stupid, and they're fighting back with their own lines of cheap relatively fast processors (Celerons), which can potentially eat away at AMD's little niche in the market.
Anyway, it would be extremely pleasant if AMD's K7 did well. When I read the false reports against their machines, I was extremely disappointed, since it would definitely nice to see AMD do well against Intel. Diversity is always good.
I remember one of the MS memos showing their concern with the high percentage of a computer price that is the OS license. I also remember one of their possible solutions was to build or buy a chip fab, and thus to lock you in at the chip level versus the OS level. Today, this is even more likely due to the shakiness of the Wintel alliance.
If they can do this, there will be serious trouble for the Linux community. One reason that Linux is so popular is that it runs on a cheap, open hardware platform.
If I were Bill Gates, I would be seriously considering a bid, and my largest concern would be sliding it past the DOJ. After all, he can put a EULA on a chip requiring you to run an MS operating system on it.
--The basis of all love is respect
Yeah, why buy a Global if you can afford a 777?
x .html
http://www.boeing.com/commercial/777family/inde
Then you can reserve an area for tennis or something. And a shower/bath afterwards.
Breace.
Well, I'm not saying Open Source chips (or Open Design), but Open Datasheets. For a start.
With datasheets I mean at least programming information, so that drivers can be written for OS's like Linux or others.
I find that a lot of the time my company (one of those small ones) is just not able to use a certain chip because we DON'T want to use M$ products (in our embedded environment, strange uh?). This means we can't use the factory M$ drivers, which means we need datasheets to write our own drivers for our own OS (or Linux).
Mainly the larger chip makers that don't have their datasheets available on the Web are a pain in the neck to get in contact with - example - NeoMagic: they don't even want to put me through to sales, let alone tech support. We don't fit in their business-plan, too small. So we use C&T because they have their datasheets readily available on the Web.
I don't believe that providing programming information for a device gives away your internal design: if I want to know the programming info real badly I will get it anyway through reverse engineering. Believe me, I have done it.
An other excuse that is used is that they don't want to 'support' the effort of third-parties writting device drivers.
Which is complete crap. If the datasheet is good, no one needs support to write a driver. For the drivers I have written (I guess about 12), I have only contacted the manufacturer twice. The first time to tell them the datasheet had errors the second time to tell them their silicon had errors.
And how difficult would it be to sell a couple of more chips to little companies like ours?
I still hope for Open Datasheets but I'm sure Open Source chips will come too. The biggest part of a chip is actually software anyways (VHDL, Verilog).
Breace.
"The few functions that made their way onto the main chip -- 3D graphics, mainly -- were put there by National's rivals: AMD with 3DNow! and Intel with Katmai."
What a load!
The MediaGX has CPU AND Memory controller AND VGA AND PCI built into one single chip. That's hardly to be called trivial and I still think that's pretty neat. A chip that comes close is the STPC from Thompson, but they don't mention that. Good journalism. NOT.
This chip is still a great device for wearables and other embedded applications. If only the support would have been better (try to find a real datasheet, or try to speak to someone) I would have probably used this chip in new designs.
(Off topic) Open Source Software is one thing, but wouldn't it be great if chip manufacturers would _finally_ be a bit more open about their designs, 'Open Datasheets'? Pffw...
Breace.
...not just Intel, AMD is a tougher competitor now more than ever. AMD has the anything-but-Intel-yet-compatible-with-Intel market pretty well cornered. Cyrix is relegated to the dirt-cheap sector, where there isn't much profit.
All the local suppliers don't carry them anymore for obvious reasons. When and what did they release last?
SP
"It is dangerous to be right when the government is wrong." - Voltaire