IDF 2006 - Conroe Merom and Woodcrest info

Page 8 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Acanthus

Lifer
Aug 28, 2001
19,915
2
76
ostif.org
Originally posted by: jones377
I don't think this has been posted yet. Real World Technologies posted a very detailed article on the Conroe Microarchitecture here.

Superficially Conroe looks alot like Yonah but when we dive deeper there are major changes. Some highlights:

* Direct L1-L1 cache transfers between the cores
* 1 Complex, 3 simple decoders. Up from 1+2
* One 128-bit FADD plus one 128-bit FMUL vector SSE instructions per cycle. Twice that of P4/K8!
* Memory disambiguation. Completely new feature to x86 MPUs AFAIK
* SSE4 CONFIRMED. Would have been included in Tejas too. Nice but not general purpose in nature so don't expect much in most cases.


Even more info in this https://www35.cplan.com/cbi_export/PS_MATS001_278703_125-1_FIN_v1.pdf">presentation</a>

id: idf
username: spring2006

good info, i hadnt heard much beyond that it was a 4 issue core.
 

pedramrezai

Member
Sep 5, 2005
59
0
0
The way most of the people are thinking is just the way Intel has planned. I am a AMD fan but I can realize competition is good for customers. I was shocked by Core performance but using a handicapped AMD system really annoyed me. First, reviewers have already proven RD580 or solutions with dual 16x can deliver up to 10-15% more performance when paired with high-end, bandwith hungry vga cards. Second, we have been hearing of dual core optimizations in display drivers for some time but were unable to see something significant until we saw Conroe performance; I am quite suspicious over some hefty optimizations in intel-cooked display driver. Time will reveal. Third, this might be the beginning of a new SSEx game with unfair optimizations for a new technoogy.
I am surprized how people are trashing the current as well as future AMD64 technology.But remember that Core is not out yet and all these might be some optimizations that has granted it this performance level. Moreover, the current AMD64 technology is almost 3 years old and the new AM2 will update its specs. AMD did not like DDR2 high latency; What they are looking for is its higher frequency that can be paired with the new AM2 FSB.For Athlon 64 and Sempron a 333mhz FSB that paires with DDR2 666 and for the Fx parts a 400 mhz FSB pairing with DDR2 800. If DDR1 could reach these frequencies you could now see the real potential of AMD64. This kind of bandwith will give Core a hard time. Also remember that AMD is increasing cache (L2 and maybe L3). Shared cache is also something that will be seen in the future products and will bring huge performance gains. Based on the preliminary data of 200/266 async single channel bandwith of 3500mb/s a memory bandwith of >10k is expected in the final product and if Intel was going to compare its future platform, it was not fare to compare it with an infrastructure of >2 years old. I am sure the new AM2 will regain AMD reputation once again. But we all must remember that this competition between major players is good for the end users.
 

Acanthus

Lifer
Aug 28, 2001
19,915
2
76
ostif.org
A64s are not bandwidth hungry...

A64s FSBs are their clockspeed...

The FEAR demo was custom and done by AT...

All of the doubts cast have been addressed in the follow up...

DDRII will not a big jump for AMD, some are predicting a performance LOSS at first.
 

jones377

Senior member
May 2, 2004
451
47
91
Originally posted by: Acanthus
A64s are not bandwidth hungry...

A64s FSBs are their clockspeed...

The FEAR demo was custom and done by AT...

All of the doubts cast have been addressed in the follow up...

DDRII will not a big jump for AMD, some are predicting a performance LOSS at first.

The question is what other improvements they bring to the table. Remember Palomino? It was more than adding SSE and upping the FSB a bit. Personally I don't expect much else besides DDR2, but some other tweaks here and there is not unresonable. They even did that for Rev E, not that it made much of a difference to performance (per clock)
 

pedramrezai

Member
Sep 5, 2005
59
0
0
A64 will be bandwith hungry when FSB is increases in parallel with mem frequency. My Sempron 2800@ 8x300=2400 paired with DDR600 memory yields a bandwith of 4.6GB/s
with single channel memory which is near the bandwith of most dual channel systems.
Using a 333 or 400 FSB and pairing them with 667 and 800 parts will significantly improve overal performance not to mention other enhancements that might show up in the final product. So DDR2 will be a real jump for AMD when FSB or whatever you like to call it is increased. In a preview system at Cebit a single channel DDR2 533 with 200mhz bus and with integrated graphics gives a respectable 3.5 GB/s. Now you can calculate how bandwith will be increased with higher FSB paired with DDR2.
 

jones377

Senior member
May 2, 2004
451
47
91
Originally posted by: pedramrezai
A64 will be bandwith hungry when FSB is increases in parallel with mem frequency. My Sempron 2800@ 8x300=2400 paired with DDR600 memory yields a bandwith of 4.6GB/s
with single channel memory which is near the bandwith of most dual channel systems.
Using a 333 or 400 FSB and pairing them with 667 and 800 parts will significantly improve overal performance not to mention other enhancements that might show up in the final product. So DDR2 will be a real jump for AMD when FSB or whatever you like to call it is increased. In a preview system at Cebit a single channel DDR2 533 with 200mhz bus and with integrated graphics gives a respectable 3.5 GB/s. Now you can calculate how bandwith will be increased with higher FSB paired with DDR2.

Dude... you simply don't know what you are talking about. The K8 never had a FSB, so by default it never had a FSB bottleneck. You seriously need to look at a K8 block diagram before continuing to spout that nonsense.
 

Acanthus

Lifer
Aug 28, 2001
19,915
2
76
ostif.org
Originally posted by: pedramrezai
A64 will be bandwith hungry when FSB is increases in parallel with mem frequency. My Sempron 2800@ 8x300=2400 paired with DDR600 memory yields a bandwith of 4.6GB/s
with single channel memory which is near the bandwith of most dual channel systems.
Using a 333 or 400 FSB and pairing them with 667 and 800 parts will significantly improve overal performance not to mention other enhancements that might show up in the final product. So DDR2 will be a real jump for AMD when FSB or whatever you like to call it is increased. In a preview system at Cebit a single channel DDR2 533 with 200mhz bus and with integrated graphics gives a respectable 3.5 GB/s. Now you can calculate how bandwith will be increased with higher FSB paired with DDR2.

sythetic benchmarks tell you what you want to hear, use real world performance not sandra. (that is if youre even talking about HTT)
 

pedramrezai

Member
Sep 5, 2005
59
0
0
Dear user, I know what I am talking about and I think your stream of thoughts needs a little bit of defragmentation so that you will be able to think more organized and to express your ideas more politely. This is my idea and I am free to express it and I think they have been based on objective documents not halucinations. If they bother you, you can simply skip them. But if you read them and you do not like them, this can not be a reason to be so angry and express your ideas like that.
 

pedramrezai

Member
Sep 5, 2005
59
0
0
I thought ScienceMark 2 is not a synthetic benchmark. Here are some other benchmarks:
3Dmark2005 cpu score=4801
super pi 1M=33s
Winrar 3.50 606 kb/s
PC mark 2002 mem score= 9224
 

jones377

Senior member
May 2, 2004
451
47
91
Ok then, explain to us where the FSB on the K8 (A64 for you) is and how it affects performance.
 

Acanthus

Lifer
Aug 28, 2001
19,915
2
76
ostif.org
Originally posted by: pedramrezai
I thought ScienceMark 2 is not a synthetic benchmark. Here are some other benchmarks:
3Dmark2005 cpu score=4801
super pi 1M=33s
Winrar 3.50 606 kb/s
PC mark 2002 mem score= 9224

winrar is the only thing in that list that isnt synthetic.
 

Acanthus

Lifer
Aug 28, 2001
19,915
2
76
ostif.org
So did it say anywhere if the dual clovertown demo was running a chipset with CSI?

I would think 8 cores would be awful on a shared FSB.
 

Cooler

Diamond Member
Mar 31, 2005
3,835
0
0
Originally posted by: Acanthus
So did it say anywhere if the dual clovertown demo was running a chipset with CSI?

I would think 8 cores would be awful on a shared FSB.

I think Its dual FSB running at 1333FSB each
 

Acanthus

Lifer
Aug 28, 2001
19,915
2
76
ostif.org
Originally posted by: Cooler
Originally posted by: Acanthus
So did it say anywhere if the dual clovertown demo was running a chipset with CSI?

I would think 8 cores would be awful on a shared FSB.

I think Its dual FSB running at 1333FSB each

Did they actually specify that they were running the dual FSB though?

Because that would mean working chipset AND cpu silicon for next gen.
 

jones377

Senior member
May 2, 2004
451
47
91
Originally posted by: Acanthus
Originally posted by: Cooler
Originally posted by: Acanthus
So did it say anywhere if the dual clovertown demo was running a chipset with CSI?

I would think 8 cores would be awful on a shared FSB.

I think Its dual FSB running at 1333FSB each

Did they actually specify that they were running the dual FSB though?

Because that would mean working chipset AND cpu silicon for next gen.


Intel already has a dual-FSB chipset called Blackford. They developed it for the P4-based Xeon systems and it should be out by now, or very soon. In addition, IBM has a simular chipset called Hurricane X3. I think the Hurricane chipset is even more advanced. Can't wait to see it combined with Woodcrest or another "Core" CPU
 

Dthom

Junior Member
May 28, 2006
21
0
0
Voted for Gore, I assume? Do you avoid chain retail such as Starbucks because they wrestle real-estate from small business owners by offering to sign super long term leases? Get with the program. Business is about making money. Intel has proven with this rapid leapfrog over AMD that they are the kings and will never allow AMD to overtake them again.

I'm of the opposite thought process than you. AMD has built their business on intel's coattails; they basically sell a copycat product and sponge into markets that Intel created; and then they sue Intel for being anti-competitive. What a joke.

Let's face it. The courts allow AMD to do what they do because there is no one else to compete with Intel. Intel is a tremendously important company that is the reason that most of you have the jobs you have, most likely. Bashing them because they make a lot money is like the poor people who sit at home at 3:45pm in their backyards complaining about how much the CEOs make, while the CEOs work until midnight. Intel has done a tremendous job in a very short time, and all of us will benefit from it. Give them some credit already.
 

Viditor

Diamond Member
Oct 25, 1999
3,290
0
0
Wow Dthom...first post ever and you use it to dig up a thread from months ago and to flame someone else without having a grasp of any facts or understanding yourself.
Could we be witnessing a new lifeform? Kind of an accelerated form of trollhood me thinks...
 

Dthom

Junior Member
May 28, 2006
21
0
0
Originally posted by: Viditor
Wow Dthom...first post ever and you use it to dig up a thread from months ago and to flame someone else without having a grasp of any facts or understanding yourself.
Could we be witnessing a new lifeform? Kind of an accelerated form of trollhood me thinks...

I don't think that someone who says he'll never buy Intel for religious reasons has much time sensitivity. I don't see many "facts" in these forums; just a bunch of back-room geeks that know *something* about computing but not much about business. i see a bunch of guys that are simply desperate for AMD to be better than intel.

Why don't you fill us in on the "facts" as you see them? People that call other people trolls without counterpointing them usually don't have any facts, because name-calling is what people who have no good arguments do to feel important.

You can spin things any way you want, but if you own AMD stock you'd best sell it soon. AMD has put all of their eggs in one basket and that basket is about to get stepped on by Godzilla. Woodcrest is only the first step.
 

OneOfTheseDays

Diamond Member
Jan 15, 2000
7,052
0
0
Originally posted by: Dthom
Originally posted by: Viditor
Wow Dthom...first post ever and you use it to dig up a thread from months ago and to flame someone else without having a grasp of any facts or understanding yourself.
Could we be witnessing a new lifeform? Kind of an accelerated form of trollhood me thinks...

I don't think that someone who says he'll never buy Intel for religious reasons has much time sensitivity. I don't see many "facts" in these forums; just a bunch of back-room geeks that know *something* about computing but not much about business. i see a bunch of guys that are simply desperate for AMD to be better than intel.

Why don't you fill us in on the "facts" as you see them? People that call other people trolls without counterpointing them usually don't have any facts, because name-calling is what people who have no good arguments do to feel important.

You can spin things any way you want, but if you own AMD stock you'd best sell it soon. AMD has put all of their eggs in one basket and that basket is about to get stepped on by Godzilla. Woodcrest is only the first step.

your a complete idiot and really should be banned. i bet you were paid by intel to come onto these forums and spout nonsense like this.

AMD's A64 was successful because it kicked the crap out of anything Intel had to offer. They held the crown for years while Intel was playing catch-up. Intel has finally put something competitive out there, but don't think AMD won't respond.
 

Viditor

Diamond Member
Oct 25, 1999
3,290
0
0
Originally posted by: Sudheer Anne
your a complete idiot and really should be banned. i bet you were paid by intel to come onto these forums and spout nonsense like this.

AMD's A64 was successful because it kicked the crap out of anything Intel had to offer. They held the crown for years while Intel was playing catch-up. Intel has finally put something competitive out there, but don't think AMD won't respond.

It sounds like Dothan (who WAS banned) with a new handle...but if we just try to avoid responses, maybe he will stop.
 

Dthom

Junior Member
May 28, 2006
21
0
0
Originally posted by: Sudheer Anne
Originally posted by: Dthom
Originally posted by: Viditor
Wow Dthom...first post ever and you use it to dig up a thread from months ago and to flame someone else without having a grasp of any facts or understanding yourself.
Could we be witnessing a new lifeform? Kind of an accelerated form of trollhood me thinks...

I don't think that someone who says he'll never buy Intel for religious reasons has much time sensitivity. I don't see many "facts" in these forums; just a bunch of back-room geeks that know *something* about computing but not much about business. i see a bunch of guys that are simply desperate for AMD to be better than intel.

Why don't you fill us in on the "facts" as you see them? People that call other people trolls without counterpointing them usually don't have any facts, because name-calling is what people who have no good arguments do to feel important.

You can spin things any way you want, but if you own AMD stock you'd best sell it soon. AMD has put all of their eggs in one basket and that basket is about to get stepped on by Godzilla. Woodcrest is only the first step.

your a complete idiot and really should be banned. i bet you were paid by intel to come onto these forums and spout nonsense like this.

AMD's A64 was successful because it kicked the crap out of anything Intel had to offer. They held the crown for years while Intel was playing catch-up. Intel has finally put something competitive out there, but don't think AMD won't respond.

Again, no arguments, just name-calling. So everyone who isn't AMD biased should be banned?

First of all, you're dead wrong. Intel was not really playing catch-up, they focused on the more lucrative notebook market, which they completely dominated and still do. The design of the Pentium-M was a substatial development, trouncing anything AMD had at the time. Then they diversivied into networking and developed WiMAX, and then, only after the market made it clear that performance per watt mattered to them, they decided to take back the high-end processor market.

Why is it nonsense? Every reputable hardware test site has essentially the same results regarding intel's woodcrest vs Opteron. Come out of your hole. Intel has 20 teams working to every one that AMD has. There's no logical reason to think that a mobilized Intel can't do things twice as fast as AMD. Intel will have 8 core processors before AMD has 4. Intel is already working on things that AMD doesn't have the resources to even think about yet. They were a sleeping giant and AMD made the mistake of throwing coconuts at them. Now they're going to pay.

My comment on AMD stock is simple; Woodcrest is faster and cheaper and uses less power than the Opteron. There will be no reason to buy an Opteron very shortly. AMD is not well diversified in its business.

Most of it is just hype anyway.I laugh heartily at the gaming sites testing high-end processors on 33Mhz bus motherboards. Most 64bit OSes I've tested are actually slower than 32bit at most things, mainly because the bigger structures cause less to be able to fit into the cache. The perfomance difference when applications fall out of the cache is significant. The truth is that most people have no clue about real-world performance. Intel has finally focused on the benchmarks, and they're going to kill AMD at their own game of hype.
 

Dthom

Junior Member
May 28, 2006
21
0
0
Originally posted by: Viditor
Originally posted by: Sudheer Anne
your a complete idiot and really should be banned. i bet you were paid by intel to come onto these forums and spout nonsense like this.

AMD's A64 was successful because it kicked the crap out of anything Intel had to offer. They held the crown for years while Intel was playing catch-up. Intel has finally put something competitive out there, but don't think AMD won't respond.

It sounds like Dothan (who WAS banned) with a new handle...but if we just try to avoid responses, maybe he will stop.


Still hoping to hear your ideas, if you have any. This sounds like a bunch of fat chicks making fun of the cheerleaders.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |