backtop


Print 48 comment(s) - last by JeffDM.. on Apr 27 at 6:46 PM


"Geneseo" accelerators are PCI-E cards
Intel claims revolutionary jump forward but takes a smaller evolutionary step first

This week during Intel's spring IDF conference in Beijing, the chip giant revealed several initiatives that it claims will give it a prominent position entering 2008. One of the most stand-out projects in the works is dubbed Geneseo, aimed at competing squarely with AMD's Torrenza technology.

Both AMD and Intel express interest in opening up their respective processor platforms to allow third-party hardware developers to produce "accelerators" that can be integrated into a system, providing specialized processing for specific tasks. Late last year, Intel responded to Torrenza during its fall IDF conference by claiming that it too will open up its server platform. Today, however, we find that while Geneseo is headed in a similar direction as AMD, there are significant differences between the two platforms.

One of the biggest announcements that AMD made about Torrenza is that it will allow companies to create accelerators that install directly into an Opteron socket -- AMD has already indicated it will be using this approach in its Fusion project. Utilizing this method, accelerators can take full advantage of the system bus as well as high-speed, low latency paths to system memory and other devices. Using Torrenza, companies can also develop accelerators that plug directly into a PCI-Express slot -- Intel's Geneseo platform only offers the later.

According to details that Intel released this week, accelerators will only connect via the PCI-Express bus. Intel claimed that this is the most cost effective way of developing an open platform.

"The majority of accelerators can be efficiently supported by PCIe," Intel revealed in its channel guidance. Currently, this statement may be true as some of the most processor intensive accelerators including graphics and specific math processors rely on PCIe.

AMD, on the other hand, states Torrenza will yield better performance by giving accelerators a direct communications path with the system processor.

Stripped of all the fancy names, Intel's Geneseo is more along the lines of an extension of the PCI-Express architecture and specification. Unlike AMD's Torrenza, Intel guidance states that Geneseo aims to simplify development time and costs -- Intel also made a note that its model is based on a more "well established compatibility model."

According to Intel roadmaps, the company hopes to deliver Geneseo systems by 2009.


Comments     Threshold


This article is over a month old, voting and posting comments is disabled

So what?
By tfk11 on 4/20/2007 2:11:44 AM , Rating: 5
We already have accelerators on the pci bus. Is there anything to this story other than a fancy marketing name for technology we've been using for years?




RE: So what?
By defter on 4/20/2007 2:39:39 AM , Rating: 1
Torrenza/Geneseo are just fancy marketing names for a faster bus for accelerators. Using HTX/PCI-E will just give you more bandwidth.


RE: So what?
By tfk11 on 4/20/2007 3:58:29 AM , Rating: 2
Exactly... I thought the whole point of Torrenza was to provide accelerators a lower latency/higher bandwidth connection than what we currenly have, which is PCI-E. Even the old PCI bus was designed with accelerators in mind (bus mastering). So how is PCI-E accelerator technology news? Am I missing something?


RE: So what?
By tfk11 on 4/20/2007 3:58:29 AM , Rating: 2
Exactly... I thought the whole point of Torrenza was to provide accelerators a lower latency/higher bandwidth connection than what we currenly have, which is PCI-E. Even the old PCI bus was designed with accelerators in mind (bus mastering). So how is PCI-E accelerator technology news? Am I missing something?


RE: So what?
By josmala on 4/20/2007 4:04:06 AM , Rating: 5
Torrenza is just not faster PCIe.
It bypasses pci protocol, and has cache coherent memory...
The cache coherence and low latencies means that it can be used with stuff that talks with processor all the time.
Its NOT about bandwith. Its all about making it reasonable for processor to give orders to accelerator, and read few values from the accelerator memory, and the give more orders to accelerator based on those values. Dependencies between reads and writes for accelerator makes this new interface needed. It doesn't make graphics accelerator stuff faster since its mostly write only device with moving of large blocks at a time. However, besides physics there are several other things that this thing could accelerate.


RE: So what?
By Justin Case on 4/22/2007 4:24:35 PM , Rating: 2
Exactly. This has nothing to do with Torrenza. In fact, the only difference between this and a classic northbridge / southbridge design is that you can hang PCIe devices (plus some mysterious "accelerators") on the northbridge instead of the southbridge.

Since there's apparently no standard for those "accelerators", I guess only Intel can make them, which means everyone else will have to use PCIe, which kind of defeats the point. It takes some load from the NB-SB link, but that's it.

The whole point of Torrenza is to give certain high-bandwidth devices (ex., GPU, vector coprocessor, PPU, etc.) direct access to the CPU(s)' hypertransport bus, _without_ going through a northbridge or a PCI layer. They can be in sockets or slots, that's not the point (they can even have their own, local, high-speed RAM). The point is to get rid of PCI's overhead (which can account for almost 20% of total bandwidth), while reducing latency and increasing the bandwidth itself.


Intel says "me too!" again
By johnsonx on 4/21/2007 12:53:24 AM , Rating: 5
While Intel certainly is leading in performance right now, don't they seem VERY "Me Too!" over the last few years? First AMD developed x86-64, and Intel said "Me Too!". AMD clearly led in developing dual core, and Intel said "Me Too!". AMD went with a lower clock speed, higher IPC architecture (as compared to NetBurst), and Intel said "Me Too!" (Core). AMD has had the on-die memory controller for years, and Intel has said "Me Too!". AMD is working on graphics integrated in the CPU die and/or package, and Intel has recently said "Me Too!". Now today, Intel has said "Me Too!" to Torrenza.

Yes, I do know that AMD had been 'Me Too'-ing Intel for years before. My AMD experience goes all the way back to the AM286. It's just interesting to see the role reversal.




RE: Intel says "me too!" again
By rbuszka on 4/21/2007 11:10:10 AM , Rating: 3
"Torrenza" is also a way cooler name than "Geneseo". The latter sounds like the name of that silly coffee maker from Philips.


RE: Intel says "me too!" again
By JeffDM on 4/27/2007 6:46:01 PM , Rating: 2
I hope it's just a codename and not the product technology name.

I thought Torrenza exploited the HyperTransport bus, I don't see how Geneseo is going to be competitive vs. just doing a PCIe card. Hypertransport would appear to remove a connection hop. I'm struggling to figure out what needs such bandwidth that it would max out an x16 slot, even graphics cards aren't using much beyond an equivalent of x8.


RE: Intel says "me too!" again
By Ringold on 4/21/2007 4:06:09 PM , Rating: 2
I agree, but I'll bring up the bane of technology leaders; first ones to market with new technology almost always get destroyed by those that come later.

Part of the reason I love AMD though is the fact that they, despite having much fewer resources, really have kept Intel playing a defensive game until only just recently. That's been a huge accomplishment.


RE: Intel says "me too!" again
By Grast on 4/23/2007 11:15:48 AM , Rating: 2
Ringhold,

I believe that most people assume that just because company A releases thier idea and development plan prior to company B. Then company A had the idea first.

1. On Die memory controller - Intel has stated numerous times in the Pentium 3,4 and no Core lineups that on-die controller would be implemented when it was needed. Intel is not copying AMD! It was a logical conclusion.

2. EM64 - I believe Intel did copy. However they copied because Itainum 64 bit pure processing did not take off like Intel expected. I believe this was mostly due to a lack of 64 bit OS from Microsoft.

Just my opinion.


torrenza socket slution not for gpu
By wetwareinterface on 4/20/2007 10:19:04 PM , Rating: 4
the point most seem to miss is that the torrenza solution is not for a desktop/gpu configuration and seem to compare it to that usage model. the point of torrenza isn't a physex chip or gpu sitting in a socket but an fpga or dsp sitting in a socket. imagine the benefits to an fpga manufacturer if the hard coding could be left out and the rom left out and the memory controller left out and the ...
basically torrenza allows manufacturers of these types of devices to concentrate on what makes their device different from a cpu and leave out all the similar circuits to a cpu and use the cpu for these tasks instead and just implement a hyper transport controller to let the cpu deal with all the little alu and memory control situations you'd need to duplicate in silicon on a fpga or dsp.

now a manufacturer can dedicate more chip real estate to parallel fp lanes or extra dsp parallelism by leaving out the parts that the cpu can do for them over there in that socket. this is about inter chip communication to a cpu to handle mundane non-specialized tasks that would get repeated on silicon in the specialized part to duplicate the same tasks like memory compares and fetches etc...

with torrenza an fpga manufacturer can get away with higher yields for the same current performance meaning higher profit. or same yield higher functioanlity higher price they can charge meaning higher profit. and let's face it getting a company onboard requires nothing more than the words you'll make more money with our approach.




By johnsonx on 4/21/2007 1:20:53 PM , Rating: 2
Ah, yes, I see what you mean. Just like an 2p opteron workstation with only 1 memory bank, the second opteron depends on the first for EVERYTHING: it's just a CPU core talking on the HT bus. Likewise a special purpose processing chip would have to implement only it's core functionality and an HT interface. Everything else would be handled by the primary CPU(s) and driver software.

While an HT interface is no trivial bit to design, it's probably already a standard building block AMD can just hand off to the chip designer.


LEADED GASOLINE IS THE ANSWER!
By wingless on 4/20/2007 7:48:16 PM , Rating: 1
We need to transition back to leaded gasoline and no less than 12:1 compression ratios and 300hp on EVERY car manufactured in the world...thatll fix the problems we have with weak ass cars and crappy acceleration.

Oh this is a debate about clean air? Cars are the LEAST OF OUR PROBLEMS when it comes to pollution. How about we concentrate on manufacturing processes, cutting down the planet's A/C(the rainforest), and the bottom trawling fishing practices that destroy the oxygen making ecosystems in our ocean. We need to concentrate on bigger problems that have conveniently been made the least visible by the media and corporations. Cars are just the most visible and easily relateable source of pollution but theyre a smaller fraction of a much bigger problem.




RE: LEADED GASOLINE IS THE ANSWER!
By wingless on 4/20/2007 7:54:32 PM , Rating: 2
ooops...

WRONG POST!


RE: LEADED GASOLINE IS THE ANSWER!
By Zoomer on 4/22/2007 9:11:15 AM , Rating: 2
Why not use diesel instead? They work well too.


Intel announces Vista SP 1
By crystal clear on 4/21/2007 1:17:31 PM , Rating: 2
quote:
Intel claims revolutionary jump forward but takes a smaller evolutionary step first


But also takes an extraordinary step by announcing the VISTA SP1.

Thank you INTEL for that early tip off.

Read this-

"In the corporate space, I believe most companies will act like Intel and do some pilots and testing today, but the deployment will actually happen when the service pack gets released in the fourth-quarter timeframe -- probably the October, November timeframe," Otellini said.

http://www.informationweek.com/news/showArticle.jh...




By crystal clear on 4/21/2007 1:35:25 PM , Rating: 2
Thank you "Otellini" for that insider information.


Folding@Home Accelerator
By Ringold on 4/21/2007 4:10:00 PM , Rating: 3
I'll take two. Thanks.




hypertransport bw more effective
By Roland702 on 4/20/2007 10:30:35 AM , Rating: 2
No doubt that with AMD's plans that having a accelerator on hyper transport will be faster then Intel's implementation. However that being said PCI-E receives and sends data at the same rate and would probably be effective with the PCI-E 2 specs.

All in all we will see the pros and cons and performance differences when some benchmarks start appearing.




R600-Tunisia
By crystal clear on 4/21/2007 11:03:50 PM , Rating: 2
AMD’s R600 media event on the 23rd and 24th April,Tunisia.

Launch date end of May-expect something from nvidia to match it (GeForce 8800 Ultra ,probably) around that time.




hm
By Russell on 4/20/07, Rating: -1
RE: hm
By defter on 4/20/2007 2:42:28 AM , Rating: 1
quote:
Of course you can't have an accelerator plug into an Intel socket until Intel adopts an integrated memory controller.


Of course it's possible, and it has been already demonstrated.

quote:
but the gignormous performance hit would defeat the purpose of having an accelerator in the first place.


Gignormous performance hit for accessing the main memory through chipset??? Certainly there is some performance hit, but it isn't that large. Just look at Core2 benchmarks...

However, this story is about external accelerators. If the accelerator needs to have lot of memory bandwidth, it can use it's own fast on-board memory.


RE: hm
By Master Kenobi (blog) on 4/20/07, Rating: 0
RE: hm
By SquidianLoveGod on 4/20/2007 8:21:53 AM , Rating: 2
But what I am worried about is, Is it going to be like the whole slot 1, slot A days? I remember so many hassles and cooling was a pain as more often than note the ram sat to near.


RE: hm
By Targon on 4/20/2007 9:18:43 AM , Rating: 2
I think you missed the big difference between the AMD and Intel ways of doing this. The AMD method provides support for both a straight processor(that goes into a socket on the motherboard), as well as the PCI Express method.

An advantage here is that if a chip goes into a socket on the motherboard, working with AMD would probably allow for a direct HyperTransport link to the other CPU(s) in the system as well as to system memory. For SOME applications, this could be a HUGE boost to performance. Also, as time goes forward, we will find that we will need a 32x PCI Express slot, which of course will require an update to the standards. With the AMD approach, you are already able to get far beyond this level of performance when it comes to a co-processor.

When it comes to other applications, where having a co-processor have it's own memory and connections, a card solution is available if it makes sense.

Looking at the AMD approach going forward, one way this could work would be to have a straight GPU that plugs into the second socket, which could make it so most high performance systems would have the second processor socket, but that it would be filled with a GPU and possibly a dedicated memory bank for memory(meaning the user could add dedicated video memory at any time). Hmmm, swap GPU chips, add video memory, and save on the PCB and other things that add to the price of a video card. Sounds like a decent solution for those who like to upgrade.

From Intel, they are saying, "We can do that too", but without providing a new way to do things. That is a part of what makes AMD different. AMD really is working hard to find ways to improve the overall design of x86 based systems while Intel is working hard to improve the different parts of a computer without looking for ways to improve the overall design. Intel has the better CPU design, but doesn't have a better system architecture at this point in time. When K10 comes out, we will see if AMD has caught up in the CPU design area. IF(and that is a big IF for many people) K10 has the design to let it compete well, even if Intel has a slight lead in processor performance, then the system architecture approach we have seen from AMD may be enough to let AMD take the lead in overall performance.

CPU+GPU on a single chip may be down the road, but CPU+GPU on a Torrenza approach may become available sooner, and may blow Intel away if cheap dual-socket motherboards become available.


RE: hm
By Master Kenobi (blog) on 4/20/07, Rating: 0
RE: hm
By Targon on 4/20/2007 11:22:39 AM , Rating: 2
This goes on the concept that there will be NO advancement going forward. In time we will see the current PCI Express bus run into limitations, and faster versions will be added.

Now, when it comes to what Intel has done, their processor design is currently better(K10 not being available), so Intel's better processor designs are more than able to compensate for a worse system architecture. This was my point, that if AMD has a processor that matches the performance of what Intel has, AMD will end up with the performance crown due to having a better system design.


RE: hm
By mars777 on 4/20/2007 10:43:13 PM , Rating: 3
I was reading till this:

"Example.
What, you wanna swap your GPU socket chip to improve performance? Oh but your board only supports 1 DVI Connection, we are sorry, you will need to purchase a board."

Man you really, and I mean REALLY, didn't understand this whole thing :)


RE: hm
By apollo7 on 4/20/2007 10:50:29 AM , Rating: 2
quote:
Looking at the AMD approach going forward, one way this could work would be to have a straight GPU that plugs into the second socket


This is exactly what I was thinking when I first read about this platform. Provided AMD comes up with some innovative GPU designs for their socket. 4X4 revised in a quad-core CPU + quad-core GPU in two sockets?

BUT wouldn't the down side of sticking a GPU and CPU in a dual socket board create a fight for system memory? Thus, we would need lots more RAM to sufficiently supply our hungry GPU + CPU.

This is just speculation on my part, so please lemme know if I'm thinking in the wrong direction!

Either way, I'm looking forward to the fruits of this platform.


RE: hm
By kalgriffen on 4/20/2007 11:27:02 AM , Rating: 2
Correct me if I'm wrong, but don't modern video cards use higher speed, lower latency memory than the DDR2 used in current desktops? Wouldn't placing a high-performance GPU in one of the sockets be offset by this slower memory (ignoring CPU/GPU memory competition for the moment)?


RE: hm
By Hawkido on 4/20/2007 12:02:24 PM , Rating: 3
Well, I see that there are many unknowns about the torrenza technology. If I were to try and put together a complete package solution using AMD's battery of technology it would go something like this.

1. Multicore CPU chip in one socket with its own bank of memory.

2. MultiGPU (or Physics & GPU) chip in another socket with its own integrated Mem controller so spec for memory can be different than the CPU socket. Allow the Torrenza sockt to use quad channel memory for higher bandwidth to the GPU chip.

3. Allocate a set of PCI-E lanes to a external slot for your <insert whatever Device interface is needed for your monitor here (HDMI/DVI-D/15-pin D-Sub/Optical)> card.

This solves most any problem you have and makes your video solution upgradable and customizable at every part. Need more video ram... upgrade your quad-channel memory for your GPU socket. need a different display interface for your new 47" optical interface monitor? Replace the interface card. Need to replace your GPU for Direct X 11 but every thing else is okay? Just pop the chip out.

I am not a system engineer, so please if there is a system engineer in the house please speak up. The rest of this is all AMD Tech that we have been talking about for a long time... I just don't see anyone putting it all together in the "Fusion" that AMD has been talking about. This is where I see it going. I think it has tremendous potential. AMD won't have to take up making RAM for videocards or buying from another fab... That will be the OEM or users job. AMD just makes the chip. AMD won't have to pay royalties for the display interface, that will be the companies that makes the interface cards responsibility.

End result: the user will only end up paying for what they need: Need Killer GPU power but not 1 gig of video memory? that can happen... Need optical video output but don't need HDMI? Well now you don't have to pay royalties on HDMI. You get the picture.

The other solution put forth by intel:
Need Killer GPU power but not 1 gig of video memory? Buy a $600 videocard that has both. Need optical video output but don't need HDMI? Buy a $800 card that has both plus 1 gig of ram and a killer GPU. Current videocard that has everything you want on it except the interface that your new monitor uses: Buy a new $600 videocard.

Which is more revolutionary? User friendly?


RE: hm
By sunbiz3000 on 4/20/2007 12:29:45 PM , Rating: 2
Perfectly said... AMD's solutions is more extensible and user friendly. Also complete customization is upto the user!! But considering Intel's dominance, its pretty tough to expect AMD's solution to win.. But lets cross our fingers and see motherboard manufacturers support the better solutions!!

Another thing to add to this is that... such sockets can be useful for Java or C# processors... You can have VMs for these languages on the accelerators, which will improve performance greatly.. Mayb latter on we can have native database instruction sets into processors.. You really can do a lot of specialization on such co-processors or accelerators as we now know!!!


RE: hm
By Targon on 4/20/2007 3:40:42 PM , Rating: 2
The big thing as far as I am concerned is to look at the overall system architecture as well as the CPU design. If AMD is able to catch up in terms of CPU design(they are behind until K10 is released in the 2nd-3rd quarter), then AMD should take the lead overall in terms of performance. People look at the current Core 2 Duo performance lead and think Intel will be able to hold the lead without doing some major amounts of work in the system architecture department.

As it stands now, an Intel based machine may have faster individual components than we had back in the 8086 days, but the overall system architecture hasn't changed all that much. You have CPU to chipset, and from chipset to expansion slots, memory, and all the other components in a system. Even with all the improvements to individual components, Intel keeps running into a wall here and there that has allowed AMD to catch up.

AMD on the other hand, is looking at how to avoid bottlenecks in overall system performance, but at the moment has fallen a fair distance behind Intel in terms of CPU performance. AMD may catch up, and Intel may pull ahead, but at some point, the AMD approach will PROBABLY allow AMD to regain a lead in terms of system performance.

This comparison is very similar to car races where you have a better driver with a car with a somewhat less powerful engine going up against a less skilled driver with a car that has more power. Depending on how things work out, one or the other will be in the lead, but in the long run, the more intelligent/skilled driver will probably win as long as the engine doesn't explode. So, AMD just needs to use a bit of nitrous to catch up to Intel. Better driving may give them the edge, or Intel's resources may let them hold a slight edge. Either way, AMD should be able to hold on and do well.


RE: hm
By defter on 4/20/2007 12:09:31 PM , Rating: 2
What is a quad core GPU?


RE: hm
By teldar on 4/20/2007 3:46:20 PM , Rating: 2
quote:
What is a quad core GPU?


That would be a 4 GPU in a BGA or LGA package which drops into a socket and has 4 individual GPU cores on it instead of one. As in lets say 4 cores with 96 stream processors in each.

Not too hard to figure out.

They just don't exist yet.

T


RE: hm
By Ringold on 4/22/2007 6:26:49 PM , Rating: 2
quote:
They just don't exist yet.


Thou hath foresaken the Voodoo 5 6000. Eternal shame must befall your GPU's.


RE: hm
By Hawkido on 4/20/2007 12:30:02 PM , Rating: 2
quote:
BUT wouldn't the down side of sticking a GPU and CPU in a dual socket board create a fight for system memory?


Check the Design specs for the Hypertransport boards... It clearly shows that each socket can have their own memory bank. And with Memory controllers integrated on each package in each socket you could run different types of memory in each seperate memory bank. No fight over memory access.


RE: hm
By defter on 4/20/2007 12:07:31 PM , Rating: 1
quote:
I think you missed the big difference between the AMD and Intel ways of doing this. The AMD method provides support for both a straight processor(that goes into a socket on the motherboard), as well as the PCI Express method.


What's the big difference? Intel has already demonstrated Xilink FPGA in CPU socket of XeonMP motherboard.

quote:
Hmmm, swap GPU chips, add video memory, and save on the PCB and other things that add to the price of a video card. Sounds like a decent solution for those who like to upgrade.


LOL. Just a quick reality check:
bandwidth of dual channel DDR2-800: 12.8GB/s
bandwidth of 8800GTX: 86.4GB/s

Even the low-end video cards, like old 7600GT, have memory bandwidth above 12.8GB/s. Your "GPU in CPU socket" would perform in ordinary GPU tasks worse than $100 add on card....

quote:
CPU+GPU on a single chip may be down the road, but CPU+GPU on a Torrenza approach may become available sooner, and may blow Intel away if cheap dual-socket motherboards become available.


may , if ... And in which market Intel will be "blown" away? Certainly not in the desktop market since Torrenza doesn't have anything to do with it....


RE: hm
By dragonlordkain on 4/20/2007 1:18:43 PM , Rating: 2
defter.. .you missed a point when the guy said that there would be a separate slot for video memory, witch could very well be a GDDR4 slot.

Now, the big spot being missed here is where people will use this stuff.
It's not a home!
This is most about server plataforms, processor clusters and all.
Yes, it is fast to have PCI-E, but it is limited in size. the GeForce 8800 GTX is already pushing the boundaries of PCI-e bandwith. Now, how do you expect to put 4 of these in comunication with 4 or 8 or 16 cores, all throught the FSB?
In the AMD case, there are already 8P boards. You could use all those spots to put processors or accelerators, all with a dedicated link with each other. No bottlenecks, thats the AMD Torreanza


RE: hm
By Nemesis 1 on 4/20/07, Rating: 0
RE: hm
By nofumble62 on 4/20/2007 10:38:44 PM , Rating: 2
High volume = lower price too. Maybe some very high end application such as specialized server or super computer can take advantage of AMD design, but the mass market will be using PCI-e because it is already existed.

This reminds me of the "PCI sound card".


RE: hm
By teldar on 4/20/2007 3:52:02 PM , Rating: 2
quote:
LOL. Just a quick reality check:
bandwidth of dual channel DDR2-800: 12.8GB/s
bandwidth of 8800GTX: 86.4GB/s


There is actually more than one type of memory. And nobody said memory controllers have to be single channel. There used to be something called a single channel memory controller. And correct me if I'm wrong, AMD bought some kind of company which uses a 512 bit ring bus. Maybe they know how to make something other than one or two channel controllers? And I think this company they bought had a design for a card, a video card I think, that uses GDDR 4? Maybe they could figure out how to share technology across divisions? Nah, they probably aren't as smart as you.

T


RE: hm
By mars777 on 4/20/2007 10:56:43 PM , Rating: 2
Best use of sarcasm i've seen this year :)


RE: hm
By wretched on 4/21/2007 12:00:14 AM , Rating: 2
Another thing worth noting is there should be nothing stopping the changes to this PCI-e standard being implemented on an AMD system, any motherboard chipset from AMD or NVida could implement the changes to the PCI-e standard and any Geneseo - PCI-e cards could then probably run on an AMD platform just as well or even better than on an intel due to pci-e to HT bridgeing, but AMD will also have the other method specifically designed to reduce the latency between co-processor and CPU.

My understanding is Geneseo is Intel's answer to try have a competitive solution quickly and bypasses some of the PCI-e overhead to reduce latency that throttles current PCI-e accelerators, and I would think it will not be able to be as optimized as the cache-coherent HT, buts its a quick patch to try have something..its a short term answer from intel, but may prove to be almost as good, but remember wont be exclusive to intel.
That said intel could use Torenza as well if they wanted to implement HT on there upcoming processors, but they want to do there own point to point bus, and no doubt will then bring out a design that competes with torenza much better, and at that stage the intel solution probably wont work on an AMD motherboard.

Remember also that HT channels can go to a 2nd socket for accelerator, also to an HTX slot that a card could be put in, and also to the motherboard chipset with when implemented with the new PCI-e standard, will give many options for the high end computing and super computing markets that want to add there beasty accelerators to systems, whether thats a chipset or PCB based accelerator, HT or PCI-e (with Geneseo methods)

The torrenza picture shown on dailytech last year helps clarify the options that an AMD system will have

http://www.dailytech.com/article.aspx?newsid=2642


Intel Genesco
By Nemesis 1 on 4/20/07, Rating: -1
RE: Intel Genesco
By Roland702 on 4/20/2007 10:48:42 AM , Rating: 2
I would seem intel these days is bent on market share these days. Their plan for a new architecture and refreshes/die shrinks also seems to work in their favor too. Though, If amd does regain market share it lost I'd like to see how prices are for it's parts.

Either way AMD would have to price competitively and so would the developers of the co-processors. I wonder how much their development costs would be too, that got to do something that will make it competitive.


RE: Intel Genesco
By decapitator666 on 4/21/2007 6:57:41 AM , Rating: 2
I would say, it shows that intels architechture is intrinsically uncapable of allowing for integrated co-processors like AMDs approach. Showing intel needs to make a radical architecture change in the future and leave what is principaly the Pentium pro bus behind for a 21 century design.
I reeks a little like a we too approach to not be completely left out in the cold.


"I'm an Internet expert too. It's all right to wire the industrial zone only, but there are many problems if other regions of the North are wired." -- North Korean Supreme Commander Kim Jong-il

Related Articles













botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki