Intel's Roadmap For the Future 135
A SV reader writes "SharkyExtreme just posted the confidential Intel desktop roadmap for CPUs. Intel is really pushing AMD with a Tulatin at 1.26GHz. and a Pentium4 at 2GHz shipping Q3 of 2001.
Also -- Intel is not abandoning RDRAM but they are adding support of DDR memory. The bottom line is that Intel is developing SDR/DDR SDRAM chipsets for future Intel processors."
Gee, I wonder... (Score:1)
Brought to you by my trustly Athlon 650.
Fawking Trolls! [slashdot.org]
Roadmap (Score:3)
--
Heh (Score:1)
aww yeah (Score:1)
The contemptuous leech Rambus is finally removed. (Score:1)
Overclocker... (Score:2)
Re:aww yeah (Score:1)
--
Sounds like Intel's really pushing Intel... (Score:2)
...considering their past problems above 1 GHz [slashdot.org]. Maybe they should get their current chips working right in quantity first, m'kay? This is more vaporware from Chipzilla - don't believe it until you see it.
Correct Name (Score:1)
Ouch! (Score:1)
'xcuse me, highly confidential? I don't think so. (Score:3)
Even if some of the details were supposed to be "highly confidential", the fact that Intel employees were giving information to what is essentially a content website erases that idea somewhat.
The information may be completely accurate or not, but it is the writer(s) at Sharkey Extreme who have put together the roadmap which we are reading, not Intel.
And we keep pedaling faster (Score:3)
Sure, we get hotter and sweatier. Sure, we burn up even more energy. Yes, we know that 13% of all electricty used in the world is for powering computers (according to a recent U.N. summit). And we all know that most of the electricity is being generated through burning fossil fuels.
Maybe we like 80-day droughts in Texas thanks to global warming.
None of that matters, though. It's too much work to move to more efficient processor designs. Why scratch our heads over big-endian/little-endian translations when we can just slap on a bigger heat sink and cooling fan? Why give up our flying toaster screen savers?
Let's stick to our 1970's-architecture processors for now. It was good enough for our pocket calculators, it will be good enough for mapping the human genome. Maybe we'll even find a way to alter our DNA so our skin is UV-resistant and our bodies need less water.
When did Intel become bad? (Score:1)
They don't have much choice but to push (Score:3)
Especially since the new mobo's supporting this are going to be arriving soon.
I'm putting off the current upgrade until I can get one of these babies.
Intel, meanwhile, has removed SMP support from the Celerons. Oops, bad timing.
Arg (Score:1)
-dk-
That name, hmm... (Score:5)
In short... (Score:1)
Remember, they won't be able to produce very many of the *huge* P4 chips on their existing 180 nanometer fabs. It'll be like the 60MHz and 66MHz Pentiums were when they first came out (to ward off the Am486), only much worse.
Re:And we keep pedaling faster (Score:5)
RISC lost the instruction set wars, however. Once it was realised that you can translate CISC into RISC using a patch of silicon, the advatanges of switching from coding in CISC to coding in RISC sort of evaporated: you *can* have your cake and eat it.
Credibility (Score:2)
I mean, sure they might release those CPUs at those given time... but they also released 1Ghz about 8 months ago and I still have troubles finding one. I don't think AMD should be really worried by that.
"When I was a little kid my mother told me not to stare into the sun...
Re:And we keep pedaling faster (Score:1)
Re:That name, hmm... (Score:2)
Re:Roadmap (Score:2)
Tom's Hardware had a nice editorial at the beginning of the year, talking about where Intel and AMD would be by the end of the year. He showed even back then, that Rambus would be a bust, and when the rest of the world caught on, Intel would try to change course... but Intel is too huge to change course that quickly.
This isn't really a plug or anything, but Tom's site was right on, 9 months later. AMD is kicking butt, their Thunderbird & Duron sets are working as promised, and Intel is still hyping vapor.
"Let's release a comparable chip speed to deflate AMD's wind", but then ship only a few hundred?
I don't care "who wins". I like the AMD prices, but have always suffered a speed hit, (I'll live). But I hate these type of marketing tactics. IBM did it back in it's waning PC days. Microsoft just did with the XBox crap. (ooh... they're coming out with a platform that will be better than the P2... in 20 months? duh.)
Oh well. Do your homework. Intel's lies are starting to show. Check out the hardcore sites that have nothing to gain or lose either way. You can't even get that kind of objectivity from the evening news.
Rader
Excellent (Score:2)
Make boards for several types and let us pick which ones work best. Maybe that would stick the proverbial fork in RDRAM.
Re:Roadmap (Score:1)
(I tried posting this already, it posted somewhere else. Let's see if it works this time.)
max speed (Score:1)
Dave
AMD to outgun Intel until Q32001? (Score:5)
AMD promised a speed step upgrade of the Athlon every 6 weeks - it started with the 1.1GHz Athlon a few weeks ago - the 1.2GHz Athlon should be released by the beginning of October. 1.3GHz in November, and 1.4GHz at the beginning of the new year. The Athlon is the PIII competitor - they have roughly the same amount of zoom in them, both 0.18micron at the moment etc. The Pentium 4, when it is released, will be a hugely expensive processor, trying to compete with Alphas from Compaq and Power3/4 processors from IBM. It is not going to compete for the desktop, corporate or home for at least a year.
So AMD will be outgunning Intel for another 6 months, possibly 9 months in terms of GHz and overall performance (ignoring the Pentium 4, as it really should figure here, and even so, the overall performance of the Athlon 1.3GHz is likely to be more than the 1.5GHz Pentium 4!)
Still, good to see Intel going with DDR SDRAM at last, and the move to 0.13micron fabs is great - although 1 taiwanese fab is already there and making stuff. AMD are going 0.15micron, probably using Motorola technology there, as the G4's were 0.15 micron...
Re:When did Intel become bad? (Score:1)
Is it any wonder they should get more slashdot atention?
Re:When did Intel become bad? (Score:1)
That said, I believe that Intel is no longer placing ID numbers on their chips, although quite honestly if they were, we most likely wouldn't know it.
stuart
Re:Credibility (Score:1)
I got a 1GHz P3 box from Dell with the 10% discount and a 2-week time to get to my doorstep. Cost me about $2k with lotsa cool stuff and a 19" monitor.
Tualatin pushing AMD? (Score:3)
AMD will be at 1.2GHz THIS year, not Q3/2001!
Weren't you paying attention? (Score:1)
GOOD: AMD, Linux(especially Debian), VA Linux, GNOME, Richard Stallman
You can not be 1337 without this philosophy.
Re:Roadmap (Score:1)
Well then maybe they should have called it Moore's Approximation or maybe Moore's Pretty Good Guess.
If you're right, I think it'll have to be repealed.
-S
Re:When did Intel become bad? (Score:1)
Intel is also the big dog and AMD is the little guy. Everyone loves to route for the little guy and see the little guy beat the big guy.
SMP vaporware (Score:2)
Re:Correct Name (Score:1)
Re:max speed (Score:1)
ROTFL (Score:1)
Re:And we keep pedaling faster (Score:1)
Re:Credibility (Score:2)
Okay so Billy Joe bob bought a 600 MHz PIII with 128MB of ram, chances are billy joe bob isnt going to NEED another system to run anything in the near future at least 3-5 years.
With SO many chips out there, most investors are making Intel and AMD just decent buys now not strong buys.
This pissing contest does create a plethora of cheap powerful chips, but if the demand starts dropping off for them (which it is predicted it will) then these companies have essentially hurt themselves.
They are right now creating a market full of bigger/faster chips and thats great but I have heard a saying work smarter, not harder and for some reason it seems to fit here.
Do more with the same, not more with more. Meaning please start redesigning processors that are more friendly, yes I know its not always possible and there is a tradeoff somewhere.. But I think the tradeoff should be a lot less.
Jeremy
Re:Correct Name (Score:1)
Yes, they do codename all their their chips after n'western cities (I'd be insulted if I lived in Mendocino or Timna BTW). Maybe it is a typo but how far back does it go? Did it happen when they founded the town, when Intel chose it as a codename, or when the Sharkster published the roadmap? Many names of Native American origin have multiple spellings -- in Wisconsin, there are at least three ways to spell "men-om-muh-nee", e.g. Menonmonie, Menominee.
--
Re:max speed (Score:1)
Let's see. I could probably use it to encode DVDs into CDROM size with divx. I could use it to decode divx and watch my DVDs. Yeah, today's higher-end PCs do this quite nicely, however DVDs and divx encoding are today's technology. There will be more CPU intensive technology tomorrow.
Likewise, there will be people with 667-MHz P3s who will want to upgrade in a few more years. They'll obviously want something a couple times faster than what they already have or they won't see the speedup they want. And if there are 2GHz platforms out there, that means that the 1.667 GHz platforms will be much more affordable for them. And who in their right mind would complain about getting a very fast machine at a low price?
If we all haven't figured it out by now, the presence of a high-end market makes everything more affordable to the remainder (probably 95%) of the market. With that in mind, I don't know what the hell people with a half a brain can possibly be bitching about when they hear about the 2GHz boxes on the horizon.
USB 2.0 Roadmap (Score:1)
ICH3 - Integrated Controller Hub 3 adds six USB 2.0 ports. From what we saw at IDF, USB 2.0 is going to revolutionize external I/O on consumer systems. We just wish we could see IEEE 1394 instead.
Intel will never support IEEE1394, and continue to push USB 2.0+ instead, because of the peer-to-peer nature of IEEE1394 - meaning you dont need a PC to connect compliant devices together. They like their own closed standard, which unfortunately is likely to become the de facto standard due with their weight behind it.
So, don't expect IEEE1394 in Intel chipsets. Ever.
CORRECTION: (Score:1)
--
Your holding the Map Upside-Down! (Score:1)
Re:Weren't you paying attention? (Score:1)
Why? (Score:2)
PS. Note that I'm talking about desktop CPUs here (which is what the roadmap is about). Servers are an entirely different matter.
Re:Credibility (Score:2)
Now to come back to my point I asked a couple of retailers in here (Montreal, Canada) and none of them can get any Ghz Pentium but all answered telling me that they could get an athlon.
Sure no one even needs that much raw power on the desktop, but god if you say you will release a CPU please release it and make it available!
We will see how things turn out but I have not been impressed by Intel lately...
"When I was a little kid my mother told me not to stare into the sun...
Re:max speed (Score:1)
Re:Correct Name (Score:1)
Re:max speed (Score:1)
Who knows, maybe a "killer app" will emerge that's completely off the radar screen right now.
By the way, back in the days of 40 and 66 MHz processors, did we ever really think that we'ed be able to fully utilize a 400 Mhz machine? Or not even fully, but be able to occassionally spike it's CPU usage up near 100%? Doubtfully...
Blah Blah, 2 GHz Q3 20001 (Score:1)
Re:Credibility (Score:1)
They're really going at it! (Score:3)
In an unprecedented move today, Intel announced that they would be taking the processor battle to another level, changing there previous policy of one press release a month to one a week. They stated that they will be announcing the release of faster processors weekly, a move that clearly has competitors worried.
In related news, AMD quickly responded by announcing a new, faster processor press release will go out twice a week, a policy that could be revised as soon as December.
Spooon!
Re:And we keep pedaling faster (Score:1)
Flames are when you attack someone directly. Flames have no supporting evidence. I presented an opinion (widely accepted, I might add) that the Intel x86 architecture is outdated. I also believe that it is inefficient, power-hungry, and its continued widespread use will worsen already bad environmental issues. I also cited a recent statement from the U.N. global conference that 13% of the world's electricity usage goes to computer systems. The fact that most of electricity is generated by burning fossil fuels is undisputed. The fact that burning fossil fuels contributes to global warming is disputed by some, but they all seem to have a vested interest in corporate industry. Their credibility is somewhat questionable.
I'm sorry you think I'm flaming. I have admittedly flamed before. However, opinions contrary to your own are not automatically flamebait.
Re:When did Intel become bad? (Score:2)
I forget the entire list of past transgressions, but I'm sure they included expensive, NDA-bound programmer's guides, the Pentium bug(s), weak real-mode/graphics support for Pentium Pro, the ID number fiasco, weak power management, lack of elegance, and lately I think its been for just not shipping 1+Ghz chips in large quantities.
But hey, it's Slashdot politics. If its not being managed by a global team of non-Americans for the betterment of Linux, there's something evil about it.
Funny: The fire department misspelled its name! (Score:1)
This is funny:
Tualatin Valley Fire & Rescue [bearcat1.com] misspelled its own name!
Proof of this is that Tualatin is spelled correctly in the medallion.
The name is Tualatin, and Sharkey's Extreme has it wrong in the early parts of the article.
Whoops. Misread that one. (Score:1)
Haha, whoops. I thought the Pentium4 2GHz would be shipping with Quake3. I gotta stop playing games as much as I do. Anyways, I'm an AMD person. I used to be 3dfx until my eyes caught on the GeForce2 GTS 64mb (Which I proudly own). Maybe the P4 will impress me enough to switch out from using AMD. Hell I went from my Celery 333 to Athlon 750. I could very well switch to Pentium4 if it's worth my money.
-PovRayMan
Re:And we keep pedaling faster (Score:1)
Re:AMD to outgun Intel until Q32001? (Score:1)
Re:Why? (Score:1)
Instead of one cutting edge chip pushing really fast (and being expensive, hot, etc.), most systems would do fine with two or more slower CPUs. The slower (older) CPUs are cheap and stable because they bleeding edge moved away from them.
People need to think about absolute speed vs. throughput. For a game you probably want absolute speed. But for word processing, web browsing, playing MP3s, and photo editing all at the same time, a SMP system composed of slower processors is a very good solution.
Operating systems and apps will have to improve to make this work. Windows 95/98/ME don't do SMP, and that is where most home users are. Running many independent apps (mp3 player, netscape, etc) concurrently will work fine, but Photoshop is one of the few examples of an app that will partition its work and use multiple CPUs well.
Re:And we keep pedaling faster (Score:1)
Could you give a reference for that? Sounds rather dubious.
You idiots - this isn't intel's road map (Score:2)
its just a roadmap, but can Intel really deliver? (Score:1)
Re:Now if only AMD had a stable chipset (Score:1)
Regarding the chipsets, VIA's struggles are laughable. The boards always fizzle under the pressure (either physically or performance-wise). They embrace such gaudy, redundant standards as DMI, ACPI, and AC97. Their 4-in-1 drivers always introduce new problems into the system, after the driver engineers just got finished with last month's bugs. Currently, there is no Linux distribution whatsoever that can run any VIA chipset 100%. Even if the system is running, the ATA chipset is running on PIO 4, and the sound system can't be initialized. What do you expect from the world's only sweatshop PCB manufacturer?
Roadmap? ...FUTURE!?!?!?!?!? (Score:1)
Like Intel is all that these days. What about an AMD Roadmap? They're the ones that have truly broken the 1Ghz barrier:
Right now the only roadmap Intell should be following is catch-up with AMD's footsteps
Re:And we keep pedaling faster (Score:1)
Remember, a huge portion of power consumers are corporate users. Computers/monitors on every desk, computers in the server room, etc.
Why would I not trust the UN? I personally don't buy into the "New World Order"/"One World Government" conspiracy. Nor am I stockpiling weapons to fight off the blue-helmeted invaders.
p3-1Ghz, 100mhz bus. Where is it ? (Score:1)
Re:And we keep pedaling faster (Score:1)
...our information economy is based on high speed computers that "gobble up 13 percent of all electricity generated...."
Facts are facts. Facts don't care what anyone believes.
Re:When did Intel become bad? (Score:1)
I guess you aren't a Linux/BSD/Unix user
Re:max speed (Score:1)
That is nowhere near true due to how this beast works. In Intels own demo of the new processor the compared a 1.5 GHz P-4 against a 750 MHz P-3.
My guess is that if you have a 6-700 MHz today you may want to upgrade to 1.6-2GHz to be able to run the next WinDoze.
Intel Plans (Score:1)
The Sharkey's Extreme article is spread out over several pages to force you to see ads which you are probably not seeing, since you have trained yourself not to look.
The following links are pages that show Intel CPU plans, and are useful for someone planning a computer purchase who wants to avoid early obsolescence. (The spaces in Page X are put in by a bug in the Slashdot code.):
Pag e 7 [sharkyextreme.com], Pag e 8 [sharkyextreme.com], Pag e 9 [sharkyextreme.com], Pa ge 10 [sharkyextreme.com], Pa ge 11 [sharkyextreme.com], Pa ge 12 [sharkyextreme.com], Pa ge 13 [sharkyextreme.com]
Those who use the Opera browser [opera.com] can load all the pages simply by holding down the control and shift keys while left-clicking on the links above.
The initial name for one of Intel's chips is Tualatin, not Tulatin. Sharkey's Extreme has it wrong in the first part of the article.
This is funny:
Tualatin Valley Fire & Rescue [bearcat1.com] misspelled its own name! Proof of this is that Tualatin is spelled correctly in the medallion.
Re:max speed (Score:1)
Re:USB 2.0 Roadmap (Score:1)
Re:And we keep pedaling faster (Score:1)
Argumentum ad verecundium. You still have yet to support the claim. Just saying,"Bob Smith said it, and he must know, because he said it on TV," isn't proof.
Yes, that's a straw man, but you still haven't proven anything.
Re:Correct Name (Score:1)
men-om-muh-nee
do do de doodo
men-om-muh-nee
do do de
(repeats indefinitely)
Seems a little biased (Score:1)
I don't watch SharkeyExtreme very often but from these two articles they seem like intel zealots who don't really look deeply into the processor problems and see what truly is going on. I don't doubt this article at all, I just question the depth of the article. I don't see any true questions raised about the chips and chipsets just repetition of how good they will be and how much they will give AMD a run.
Yhcrana
Re:When did Intel become bad? (Score:1)
1. Total domination of motherboard chipset market. At first (before Slot 1) nobody cared because you could run your AMD and Cyrix chips on Intel chipsets.
2. Slot 1. Now Intel makes their bus protocol proprietary. Meaning AMD and Cyrix can't stick their chips on Pentium 2 motherboards. And because of (1), they must now depend on inferior 3rd party chipsets.
3. RAMBUS. Intel pushing slower, proprietary, and expensive memory to get lucrative shares in the company. And then not providing quality SDRAM mobo chipsets after 440BX to force use of RAMBUS.
4. Demanding an injunction against VIA to stop all imports of their chipsets. There would then be no Athlon motherboard chipset.
5. Possible: www.faceintel.com . But perhaps not enough people know about it to make an impact.
6. Probably other stuff too =P
Re:'xcuse me, highly confidential? I don't think s (Score:1)
Re:And we keep pedaling faster (Score:1)
Re:When did Intel become bad? (Score:1)
RISC vs. CISC: the Post-RISC Era (Score:2)
RISC vs. CISC: the Post-RISC Era
http://www.arstechnica.com/cpu/4q99/risc-cisc/r
(please support my karma whoring...slashdot is eating my karma for some reason
And Intel... (Score:1)
Re:Why? (Score:1)
I would guess that it is probably about as cheap for Intel to fabricate 600MHz Celerons as it was to fab 400MHz versions six months ago.
The price stays the same but the power increases.
Was it you a year ago crying that there was no market for 1GHz chips? You'd be foolish to say that now. Excuse me now, I'm going to go play Quake III at 1024x768 on HIGH while downloading game demos using Getright and keeping Netscape and ICQ open for the convenience.
Re:Credibility (Score:1)
Re:max speed (Score:1)
What I would like to see is faster buses (regular PCI is too slow for say a nice SCSI controller), better HD interface (IDE sucks), and (gasp!) DDR memory. But that is not Intel's agenda.
___
Re:And we keep pedaling faster (Score:2)
Here is a pretty good (though incomplete) explanation why modern x86 processors are not equivalent to RISC processors:
http://www.realworldtech.com/page.cfm?ArticleID
Stop spreading 5 year old Intel marketing lies.
Dell got it (Score:1)
___
Re:SMP vaporware (Score:1)
"A couple of other interesting things from the conference. According to the technical presentation the multi-processor board will be able to support only 2 processors on the Northbridge. We were told that the current Athlon chip has no issues with multi-processor functionality, the issue lies in the way that the Northbridge chipset was designed. Also, according to the tech guy the multi-processor boards will be able to use processors with differing speed grades (i.e. a 700MHz and a 900MHz processor running on the same board simultaneously). AMD will use the LDT bus to connect multiple Northbridge chipsets to allow multi-processing with more than 2 processors."
"As was pointed out previously, the North Bridge only supports 2 CPU's, however, the LDT bus supports multiple North Bridges, so 4, 8, and 16 CPU (and theoretically more) configs are possible one the LDT bus chipsets come out, mid 2001..."
"The LDT bus runs at 800Mhz, and 1.6Ghz clock rates, which means it can move 6.4 Gigabytes of data per second each way, 48 times the bandwidth of, and at a lower latency then the current PCI bus..."
"Based on their Diagrams, the CPUs' will communicate to the 'memory hub' across the EV-6 bus (at 266Mhz, moving to 400Mhz eventually), and the memory hub will communicate to peripherals across the LDT bus, which includes the I/O Hub, at 800Mhz (moving to 1.6ghz eventually)"
Re:When did Intel become bad? (Score:1)
Re:When did Intel become bad? (Score:2)
Not a confidential Intel Document (Score:2)
#2. Even if it was direct from the horses mouth (or Intel's) it would not be meaningful except historically. Intel has been lately forced to rush their chips to match market forces where for years they were able to determine their pace of innovation on their own. As such we have watched them flounder.
#3. As to WHO needs 2ghz chips, who needs more than 640k ram? In 5 years we have to assume that applicaitons will be much more involved on some levels. The idea of wordprocessing might be dead by then, with dictation as the primary purpose. Multimedia interactive reports might be the norm, and pure text might be considered retro...
#4. Multiple Processors would be nice if we had truly modular computers. I would love to see a computer which you can just snap together from parts and increase in ram, processors, storage, and such without any limitation, and without turning off machines. Anyone wanna give me some money to create it?
Re:That name, hmm... (Score:2)
And for the record, Willamette is pronounced wi-LAM-met, not wi-lam-MET. Don't ask me why; I'm originally from Tennessee and I think Oregonians talk funny, too. ;-)
--
Re:Roadmap (Score:2)
It's not like they actually get paid for each page download, is it?
The webmasters of these hardware review sites would do themselves a favour if they tested their designs. I seriously doubt they're getting more click-throughs with this mini-page design. If they are, well, then, I'm just weird.
Last time I was at Anandtech, I was able to use a dropdown menu to leap to the concluding remarks. I loved that: cut to the chase; if it turns out I'm interested, I'll go back read the rest of the review.
When I come across a design like Sharkey is using, I tend to give up before I get to the end. I know that someone else will publish much the same information, or will convienently summarize it for me in a weblog.
Sharkey, are we there yet?
--
Re:100% compatible? No, and I have proof! (Score:2)
As for the chipset, my 440BX is at about 90F. The DDR RAM chips on my GeForce 2 are usually about 110F. My P3 is about the same. I dare you to stick your finger on the 750 chipset and the CPU itself. Chances are, you can't even hold it on there for more than 300 milliseconds before getting scorched by the 180F heat.
As for the boards, at least mine wasn't built in a sweatshop. Speaking of foreign, I'm guessing that you are of either German or Russian nationality. Considering that AMD is funded by the German government, the very fact that you step right up to their defense solidifies my theory. "They are %100 percent compatible." Isn't that a little redundant?
Anyone else worried about radiation? (Score:2)
Re:Why? (Score:2)
Re:Anyone else worried about radiation? (Score:2)
Re:That name, hmm... (Score:2)
Well, I took advantage of the opportunity and ran with it. It would be a very fitting name anyway.
> [...] and I think Oregonians talk funny, too.
He, he, and that coming from a Southerner. Go figger...
Re:Why? (Score:2)
Would I? I'll never claim that there's no market for high performance chips, but I still can't see a mass market demand for 1GHz CPUs. Yes, Intel/AMD's marketing depts may be able to convince the world that there's a need, and thus artifically create a demand, but in reality, very few people have a need for chips that fast.
Re:damn hardware review sites (Score:2)