Can AMD survive Bulldozer's disappointing debut?

[rich wargo]

Originally Posted by mmarq

You may got a point for now.. but even the same exact chip and 4.2Ghz base is possible, and i'm pretty convinced that will happen before the end of summer next year.. then your words can be completely extemporaneous.. but for this we have to ask our friend rich here for another push also...

If'n I push any harder, the damn building fall over.

[mmarq]

Not that it matters that much.. but thanks anyway rich, i know i can count on you for not let me look like a fool in those predictions of mine lol

@system7

This might interest you... its about that OS scheduler, the CMT thing, and its not exactly the problem that is spinning around, actually 1 heavy thread/process and 1 low load low priority thread can co-exist in the same module quite nicely.

Windows8 brings close to 10% improvement compared to windows 7... i'm sure Linux can do better.

http://www.tgdaily.com/opinion/59039...s-of-windows-8

The processor's benchmarks improved dramatically when running Windows 8, compared to its performance on Windows on the aging Windows 7. Now I was expecting a lot of things with Windows 8, say like faster boot times, the new Metro interface, and a new suite of applications, but I certainly wasn’t expected a massive increase in performance.

I'll be damned !... has TG actually gone 'play' to AMD side ? ... dramatically ? massive ? ... "drama !" ... isn't that a charm for anxious users, what could they love better ?

nevertheless 7-8% is already quite good for today standards.

Now it only needs speed... and we can count on rich ... ummm... 4.2Ghz please? thank you ! lol

[mmarq]

i missed this one...

Originally Posted by Exophase

HTX3 and PCIe v3 serve different market needs. Even the HTC slides demonstrate this. The former is really not poised to replace the latter. And those market needs are mostly limited to HPC. Consumer graphics cards running games aren't limited by PCI bandwidth and they certainly aren't limited by latency. Just look at the tests with using 3x SLI over three 16x PCIe slots vs three 8x ones.. that's double the bandwidth, and how much of a difference does it make? And it seems to me that even if HTX did offer improvements in the desktop market it's moot when AMD's desktop CPUs don't actually provide additional HT links (outside of the southbridge one), unless I'm misunderstanding exactly how this works..

Yes many of this things you say are true... but could a GPGPU card be only for games on the desktop, or something more?... what about something like raytracing in games(primary rays illumination) ?... and what about cache coherency for this ?... what about the good power management that HT has, ganging un-ganging lanes with turbo could be implemented ... actually HTX3 is old tech by now, HT consort. could had gone for HT 4.0 and HTX4.0 by now, they have already in papers from contributers quite interesting propositions...

... it could have more speed -> even less latency, it could have content aware routing, lightweight notification, compatibility with the new multiplex scheme of PCIev3s(search for all of it)... making xfire scale tremendously better than now, with less bloated software drivers... and coherency is perfect for the kinds of GNC and make those cards like full partners in co-processor orientation, meaning a graph card would also be like a DRAM expansion...

None of it PCIe can provide i'm afraid.

[Guild]

With Windows 8, hopefully more multi-threaded apps, fixing BD, and IVB offering similar performance to SB, whatever comes next with PD cores will do better in 2012. It should be faster than i7 3770K, at least in multi-threaded apps. Albeit at 60% more power... But for many price/performance is a more important factor anyway. Still I think the market for FX is becoming more and more limited. The volume is in Trinity and Brazos, the money in Opteron.

[Jasn]

Originally Posted by Guild

With Windows 8, hopefully more multi-threaded apps

With Windoze 8, unfortunally more C#\Silverlight apps. Optimization is not hip these days, it's all about yOzEr XPirienZe. Sorry, couldn't resist.

[carop]

Originally Posted by Exophase

Still, if AMD really does become more competitive against nVidia in the HPC world, which is absolutely their intention, I don't see what market forces would be preventing them from releasing an HTX version of their high-end discrete GPUs. That is, if it really does provide a tangible enough improvement in markets where it's relevant.

Cray will probably be one of the factors for using PCIe.

Cray's next-generation Cascade supercomputer is scheduled to be introduced sometime between 2012 and 2014. It is partially funded by the High Productivity Computing Systems program of DARPA.

One of the key components of Cascade is a new interconnect code-named Aries. Aries replaces the HT links of the current Gemini interconnect with PCIe links. This suggests that future AMD Opteron and Intel Xeon processors as well as GPU co-processors will have embedded PCIe circuits.

For example, Cray has signed a multi-year, multi-phase contract with the High Performance Computing Center at the University of Stuttgart. The delivery of the Cascade system, is expected to be completed in the second half of 2013:

http://www.prace-ri.eu/More-capacity...-PRACE?lang=en

There is also some discussion that the Intel MIC architecture could use QPI rather than PCIe. It is possible that they will have QPI variants. The MIC architecture targets the HPC segments, and it seems that Cray is the major factor for using PCIe instead of QPI.

[polenta]

http://guru3d.com/news/intel-8core-s...ep-benchmarks/
Cinebench 11.5
Now, this is depressing.

[carop]

Originally Posted by polenta

http://guru3d.com/news/intel-8core-s...ep-benchmarks/
Cinebench 11.5
Now, this is depressing.

The numbers are from the usual suspect. They should have given him credit for those numbers. He does seem to have the silicon. The last time I heard about this CPU, it was on sale at e-bay.

It seems Intel is not without problems. There are problems with the virtualization engine and SAS ports. However, the real problem is with the power that is required by Sandy Bridge-EP. The clock rate of the high end workstation model will reach 3100 MHz with a very generous 150 TDP. I think that is a lot of power for a CPU to be acceptable:

http://www.cpu-world.com/news_2011/2...ries_CPUs.html

When I look at the table above, I see that most of the Sandy Bridge-EP Xeons will not run above 2500 MHz. They may make sense as Nehalem replacements from 3 years ago, but I would rather wait and see how they stack up against the Westmere-EP.

If only AMD could have delivered "35% more performance with 33% more cores" ...

[rich wargo]

Originally Posted by mmarq

Windows8 brings close to 10% improvement compared to windows 7... i'm sure Linux can do better.

Windows 7? Windows 8? What about Windows XP?!? I still depend on that day in and day out, especially at work.

Originally Posted by mmarq

Now it only needs speed... and we can count on rich ... ummm... 4.2Ghz please? thank you ! lol

Working on it, mmarq! Hey, Tim, get da axe and chop dem timin' crystals in half, we need ta double da speed on dem damn chips!

[rich wargo]

Originally Posted by carop

Cray will probably be one of the factors for using PCIe.
.
.
.
Cray's next-generation Cascade supercomputer is scheduled to be introduced sometime between 2012 and 2014. It is partially funded by the High Productivity Computing Systems program of DARPA.

There is also some discussion that the Intel MIC architecture could use QPI rather than PCIe. It is possible that they will have QPI variants. The MIC architecture targets the HPC segments, and it seems that Cray is the major factor for using PCIe instead of QPI.

Cray would make an interconnect out of paper mache if someone would pay for it. They are government money hos.

Don't read anything into this, except that Intel is supplying the chips and Intel mandates PCIe.