Sign in with
Sign up | Sign in
Your question

Intel's New Roadmap

Last response: in CPUs
Share
December 4, 2005 2:01:35 AM

http://www.tomshardware.com/cpu/20051203/index.html

Well, the future looks promising. It looks like Intel is preparing to flood the market with new processors.

What's most interesting is that they already have the quad-core Kentsfield already tapped out. And the Kentsfield is for the consumer desktop market, not the server one. They also have 8-core processors set for 2008.

45nm is set for at least mid-2007 although I'm still hoping for earlier based on their earlier statements.

I guess Intel is finally responding to AMD. Competition is great.

More about : intel roadmap

December 4, 2005 5:20:10 AM

Quote:
It looks like Intel is preparing to flood the market with new processors.

Just what we need. Hopefully Intel will be able to confuse enough people to be considered competition for Amd. Then Amd might actually do something usefull.
If Jonah2 starts @ a faster speed than Amd can muster, it may take the crown. Fact is, it's not as good as a turion, and uses more power. It really is a pity that no-one has come up with a decent platform for turion. Probably afraid of an Intel response.
Related resources
Can't find your answer ? Ask !
December 4, 2005 8:32:49 AM

News Flash ltcommander_data anyone can read tomshardware main page.
December 4, 2005 8:38:03 AM

:roll: It still warrants a discussion.
December 4, 2005 1:57:25 PM

Quote:
...change the simple fact that performance decides over victory of defeat


Cool!

I didn't know George W. Bush wrote for Tom's Hardware!
December 4, 2005 2:14:18 PM

What's good with road map is that nobody can say if it will be real or if it will change in 6 month. While this road map is full of nice promises, I remember all the nice promises that Intel gave us with the P4 that should go up to 10 GHz, or with the 90nm that were supposed to be cool and performing good.
The A64 from AMD has been delayed many time and late according to the AMD road map. What about M2.. pushed back some more... And what about ATI and their chipset or GPU??? Seriously, don't say me you believe road map!!!

And, while I'm not an ingeneer, I don't need a road map to see what's coming next. Look, we all know that shrinking the die and smaller wafer cost less to produce. So, I'm sure that both Intel and AMD are working toward that way. It don't matter which one will get there first, what matter is which one will have the best implementation of that 45nm.. Remember the Prescott. First in front of AMD, but rather deceiving in term of overall performance. Then, with heat and power consumption becomming something more people care about (the reason I have AMD right now.. cool and silent computer with any fancy cooling), it is clear that both will work toward this way. There again, someone will have an advantage. Who? I don't care right now, only when I'll be upgrading. Timing is something that matter here. Because we all know that with every revision, cpu improve (the first Prescott were a disaster, newer are easier to manage and newer AMD revision E just improve the already good Venice core. If you don't know yet about Rev E AMD 64, then you are not really CPU expert, but only tech follower..). So, who will be eager to release a newer thing, even if not ready, first and later improve it, risking a disaster will have to be seen. Now that dual core are there for good, who will have an edge at that will remain to be see. While Intel looks good on paper (anybody that had a blind date with someone that looks good on description, but not really once there can approve that), the technology is still on paper. Same thing for AMD. Their first X2 were a succes, because there was no really competition in term of performance from Intel, apart from the price performance. (But I'm sure that Intel is not happy to have to price these CPU cheaper than AMD, believe me). But how do they will stand against the newer Intel. Their road map looks promising too...but it is still something put on paper.

But anyway. I don't know why people get so emotive about a brand. While I use AMD right now, I actually buy my CPU and I expect them to deliver quality and performance at the price I'm willing to pay. If I buy Intel, I expect the same. I don't get money from AMD to use their CPU, nor I will receive from Intel if I ever use them. But one thing is sure. By being tied emotionally to a compagny, even if it don't do good, then you are the one that will be miserable. Because you won't get the true value for your hard earned cash. Don't dream for the future.. live now and enjoy what is there. because the nice CPU you are dreaming about right now and supposed to be there by half 2006 migh be either late or discontinued, because it won't be valuable enough for the mass.. they won't make a CPU just for you...

You may have technology, but if no one want it or don't want to pay for, then it won't do good. Ask RAMBUS. Why do DDR succeed? because it was memory for the mass instead of being a proprietary thing for a smal bunch of user. DDR2 might seems better that DDR just because it has a 2, but it is still only a way to store information.. By the way, I had a smile about what THg wrote"" if AMD can use DDR2 to its potential", or something like that.. My guess is, if they can design CPU, with much more precise timing and logic inside their CPU with succes, my bet is they will be able to implemente the loosy timing of DDR2.. But, they had to make AMD look bad in this article too... So they will have their money from Intel.

The best road map to me, is to look at what is on the shelf right now and decide which one give me the more for my money. And right now, Intel current offer simply sucks and AMD rulez[/fanboy]. tomorrow? I don't care.. I'll see when I'll be there.
December 4, 2005 3:58:08 PM

I don't think that these new processors will be that confusing at all. What I meant was that Intel is actually introducing new processors across the board. With Conroe replacing the 9xx, Allendale replacing the 6xx and Millville replacing the 4xx. The most interesting is Millville since with 1MB of cache and based on Merom, it should finally bring significant performance increases to the Celeron D series.

Somehow I doubt that a dual-core Turion will use less power than Yonah2. A single-core Turion produced on the 90nm SOI process uses similar power to a Dothan. Yet if AMD hopes to field dual core competition to the Yonah2 in the notebook market they would have to introduce the new Turion in H1 2006 meaning it would still be using the 90nm SOI process. That means that while Yonah2 has the shift from 90nm to 65nm to buffer the power transition from single core to dual core, Turion would have no such luxury. If AMD were to wait for 65nm then that means that a dual core Turion won't be released until H2 2006. This would give Intel 6 months lead in the dual core notebook segment and a dual core Turion would be competing with Merom.

Either way, a dual core Turion needs to wait for the S1 Socket which isn't available until Q2 2006. This means Yonah2 which is released on January 1, 2006 will have at least a 3 month lead. The 2.33GHz Yonah2 is also scheduled for released in Q2.
December 4, 2005 4:53:36 PM

I'm not saying I believe everything the road map saids as a certainty. Of course not, that would defy the meaning of a road map. I'm just pointing out that Intel finally has a decent plan for the future that doesn't include those 10Ghz P4s that you mentioned.

In any case, what I'm most concerned with is the near-term of that road map. And the near-term looks even more promising than what that road map saids. For example it saids that Presler is to be launched early 2006. In fact it the 955EE launches December 27, 2005, with the rest coming January 15, 2006. Similarly early 2006 for Yonah2 means New Years day. As well Dempsey is listed as being mid-2006 while in reality it is being pushed forward like Presler in order to correct the mistake of Paxville. The performance numbers are already out and a Q1 launch is more likely.

What's most interesting is that while Merom and family is listed as End 2006, a Q3 launch is more likely. The Inquirer has already stated that Merom is way ahead of schedule and its been taped out and in production since June of this year.

While I share your reservations about the loong-term viability of road maps, the processors listed for 2006 are believable and are in fact ahead of schedule. Besides, road maps try to incorporate buffer room to account for possible problems and to allow them to pat themselves on the back when they beat targets. For instance, the quad-core Kentsfield which is meant for the consumer desktop market is already taped out with prototypes in production while its launch date isn't until mid-2007.

And in regards to Tom's AMD DDR2 comments, I didn't find them that bad in context.

"Then there is the obvious question of how Intel's new technology is going to influence the current balance of power. We can't say yet, because there are too many variables that need to be considered. Will AMD be able to take advantage of DDR2 memory?"

They were just asking whether Merom and family will be successful, and saying that AMD's DDR2 implementation is a determining factor in that. That seems perfectly reasonable.

Besdies, the initial M2 processors are still K8-based. K8L is scheduled for 2007. That means they aren't redesigning the internal workings of the processor for the initial M2 socket. The only difference is the DDR2 memory controller. One way or another, the latency will still increase by going to DDR2. Their DDR1 controller was ultra-efficient and I have no doubt that the DDR2 controller will likewise be so. Cancelling the controllers out will still yield the increased latency going from DDR1 to DDR2. The thing that offsets this is of course the increased bandwidth of DDR2.

About Intel pricing its processors below AMD. I don't think its really that big an issue anymore. Going from 90nm to 65nm would have decreased their production costs especially since Cedar Mill is identical to Prescott2M. They wouldn't be losing profits by aggressively pricing since they're just passing on the savings to consumers. The situation is even better in the chipset market. Currently chipsets are produced in 130nm using 200mm wafers. They are now transitioning to 90nm on 300mm wafers.Going from 200mm wafers to 300mm wafers alone increases the number of printed dies by 240%. That isn't even including the transition from 130nm to 90nm. Intel could easily drop their chipset prices in half and still increase their profit margins.

http://www.xbitlabs.com/news/chipsets/display/200511242...
December 4, 2005 5:20:47 PM

That whole DDR2 thing has something that I don't like for now.. Latency.

Just imagine a road with a certain speed. then authorities declare:Éwe will up the speed limits, but we'll add stop sign and lights.... At the end, does this can be called an improvement?

Having bandwidth is good, but do you need all of it? Intel CPU do, because of their design. AMD don't, because of their design. So, do more bandwidth will help AMD? If the CPU don't need it, it won't matter. Even, if the internal memory controller prove to be efficient in cutting those latency, then, unless they end up with multicore and bigger cache, DDR2 won't really improve thing. But they won't make them badder anyway. So, I just see that like a move to follow a current trend. Intel has a lot to win with DDR2 otoh. They are bandwidth hungry and DDR2 could be one of their weakest link. But I guess that as DDR2 memory will improve, so will the memory controller in their chipset, and third party one.

The newer Opteron socket 939 can be OCed on air at 3.0Ghz. New A64 rev E will be probably able to do the same. What that means is that it may help AMD to keep up with Intel for a while by increasing MHz. Maybe it is one reason that they pushed back the M2 release with newer core design.

Anyway, I'd rather wait to see what's will be next. Because even if all that looks good on paper, they are not on the shelf right. I'm getting a new CPU soon. That will be the newe A64 revE that i will have fun with for a while. After that, I'll go dual core. right now, AMD offer is too expensive and Intel not enough performing for simple apps.. When new product will be available and tested, then I'll start to care about my next big upgrade. But I rather hope for better chipset with faster subsystem and faster HDD.

I'd rather see a new connection standard way faster than USB or firewire while maintening hot plugging. The technology for that is there. this would bring faster external storage devices and better video capture external unit. Brign HDD that can throw out double of what they can do now and you see that this will increase overall speed of your system much more than a new CPU or memory.
December 4, 2005 6:08:48 PM

AMD was kind of forced into adopting DDR2, since production has pretty much switched over. I think DDR2 is actually cheaper than DDR1 now. DDR2 won't benefit AMD's single-core processors since as you say they don't need the bandwidth and the increased latency may actually hurt them, but I'm sure it'll benefit their dual-cores.

One thing about Intel is that they design and manufacture their own chipsets. If they really wanted to they could invest the money to really reduce the latency of their northbridge memory controller. They could easily optimize the data pathways between the processor, the northbridge and the RAM. It won't be as good as AMD's integrated memory controller, but it'll certainly see some improvement. I think those early Dempsey reviews mentioned that the memory controller seemed more efficient.

For external hard drives, the future is supposed to be external SATA although I haven't actually seen any at retail. It offers the speed, but personally I don't ever see it catching on becuase its jusst not common enough or compatible. Firewire 800 is pretty good with its 100MB/s potential, and its backup compatible with Firewire 400 which is becoming more common now.

I really don't see how they are going to increase hard drive performance. Adding more cache helps with bursting but it only goes so far. I thought they were planning on caching the entire OS startup routine so that boot-up could be quick. In the end, HDD are just too physically limited. I wonder if perpendicular recording is any faster than what we have now?
December 5, 2005 1:49:10 AM

Quote:
Besdies, the initial M2 processors are still K8-based. K8L is scheduled for 2007. That means they aren't redesigning the internal workings of the processor for the initial M2 socket. The only difference is the DDR2 memory controller.


Do you really know what you're talking about???

So, extending the AMD64 instruction set and incorporating new multimedia and FP instructions are not a great redesign?? 8O

AMD's GREAT redesign will be their new architecture scheduled for 2008 (some might call it K10).

Keep trying... :wink:
December 5, 2005 3:43:03 AM

The article dabbled with incoherence at the best of times, slaughtered grammar, knifed spelling in the belly and was a slather of broken English which left me little the wiser.

Mostly it was a string of codenames, which no doubt will be changed by marketing closer to the shipping date, and told me nothing of the relative capacity or performances of the chips.

The very things I think most users would be interested in.
December 5, 2005 3:48:34 AM

Yes, the Intel spin team has done thier job once again.
There is more substance to what they are saying, perhaps, than usual, but still no where near the spin that Williamette had.
December 5, 2005 4:28:50 AM

What can you expect when they're coming out in Q306? :roll:
December 5, 2005 4:42:15 AM

The moving finger writes, and having writ moves on
And not all thy piety nor wit can lure it back
To cancel half a line
Nor all thy tears wash out a word of it.
December 5, 2005 4:42:35 AM

Then it is no better than the petty stories people salivate over on E!

"Dressed in a slinky black number by Franks & Corgette, the CEO of Intel denied today he was expecting in the new year... "
December 5, 2005 4:52:39 AM

No better then the stoies on E? Bad example.
December 5, 2005 6:46:44 AM

roadmaps subject to change at any time. believe them when you actually see the products in the store.
December 5, 2005 8:37:41 AM

Oh I agree.
December 5, 2005 10:26:48 AM

I found this great analysis here

Give it a good read. :wink:
December 5, 2005 10:40:47 AM

Pretty crappy but then I'd expect that from amdzone.
!