Conroe 2.4Ghz on 965G mobo, brief test...

[vanovich]

#892 sounds great .so lets see if it can booot

[Plaicd]

I've managed to get one. Should have it this weekend...

Retail?

[integral]

I've managed to get one. Should have it this weekend...

Are you sure it supports Conroe?

[TarTheDark]

"Some sources in the industry claim that Intel’s processors code-named Conroe have higher performance compared to AMD’s solutions, which is why a mainboard that supports Conroe may be more attractive to enthusiasts compared to platforms for AMD microprocessors. While Albatron is vague about Conroe support by PX975X, the insiders within the company said that the new chip will in fact be supported."

Source: http://www.xbitlabs.com/articles/edi...t2006-3_5.html

[integral]

"Some sources in the industry claim that Intel’s processors code-named Conroe have higher performance compared to AMD’s solutions, which is why a mainboard that supports Conroe may be more attractive to enthusiasts compared to platforms for AMD microprocessors. While Albatron is vague about Conroe support by PX975X, the insiders within the company said that the new chip will in fact be supported."

Source: http://www.xbitlabs.com/articles/edi...t2006-3_5.html

I've brought this and another link to this thread. Question is whether it is sure or only rumours.

[freecableguy]

Retail?

No. not retail.

[Iconyu]

Any idea of when it will go retail?

[Carfax]

Victor, have you confirmed what stepping the chip is?

BTW, I asked about the Sciencemarks scores over at Aces, and Redpriest (one of the guys who works on Sciencemark) had this to say:

Conroe's score is amazing.

Keep in mind that he tested both the Pentium-M optimized binary and the Pentium 4 binary. Out of both binaries, the Pentium 4 was faster. This is compiled with Intel's latest *publicly* available compiler.

Conroe is more similar to the Pentium-M than to the Pentium 4, but the binary doesn't utilize Conroe's wider resources. In *scalar* code, x87 code is generated by the Pentium 4 optimized binary because the Pentium 4 can't execute pipelined scalar sse adds, but can in x87 mode. (I had thought this was not the case but Intel's optimization guide says this is the case, I'm not sure if I 100% believe that, but given that x87 code was generated with the most aggressive flags possible -- sse/sse2 code was used sparingly in with some of the x87 code).

I imagine that if you ran the 64-bit binary Conroe's lead would widen even more.

What I am perplexed about is why Conroe bombs on the encryption code -- the entire instruction mix consists of BSWAP, XOR, and various MOV instructions, none of which are micro-coded on other processors, and no jump instructions. With Conroe's wide integer resources, I don't see why this is the case. I'm guessing there is an address generation limitation, but since I don't have a Conroe I can't really test that theory. It's a blind guess.

In any case, I see this as a *strong* showing for Conroe, not a weak one as the blogger claims. In the benchmarks that matter (BLAS, MolDyn, Primordia) Conroe is at least equal to if not exceeding an Athlon at the same clock speed. The fact that in 32-bit mode, an Athlon64 clocked 400 mhz higher cannot exceed Conroe's performance gap is telling.

I can't wait for a 64-bit run....

A good explanation, and depending on what stepping Victor is using, it may explain the other anomalies.

[Fred_Pohl]

No he doesn't.

Overclockers with normal sense wouldn't sell the skin of the bear before it's even shotten.

I see the half of this thread saying amd am2 got owned?
may i ask where i already can buy this and can you give me proof of this?
I don't give a about amd getting pwned or not..
But so far nobody is getting pwned except the previous generation of cpu wich is already there for years..

Haven't you seen any of the AM2 performance previews? I don't think the 2-3% better performance of AM2/DDR2-800 over S939/DDR400 is going to be quite enough to hang with Conroe. Do you?

Also, how many years have dual-core K8s been available now?

[Fred_Pohl]

;D

I badly need a new cpu to play with. To bad Conroe won't be available for 6 months. I have a feeling AMD just dropped the prices on the Opterons to get them selling again just like before christmas. Otherwise, they could sit on a lot of silicon soon.

6 months? Where did you hear that? Most sources are indicating July or August Conroe availability. You can pay $1000 now for a FX60, $1200 in June for a FX62 or $300 for a FX62 spanking E6600 in July or August. What makes more sense to you?

[redpriest]

ScienceMark was not written to show the power of the Athlon -- ScienceMark has been around for a while, in the 1.0 incarnation and in the 2.0 incarnation; and we receive no help from either AMD or Intel in regards to the benchmark. Any software or hardware is bought out of our own pocket -- I have bought numerous Pentium 4 machines and Pentium M machines, and I imagine when Conroe becomes available for me to purchase, I will purchase a Conroe as well.

Carfax posted my reply over at Aces' (thanks Carfax for posting it here) -- Intel actually recommends using x87 in scalar situations over SSE on the Pentium 4, due to x87 adds being pipelined. 64-bit code will probably be

Conroe shines in ScienceMark because it is a *strong* architecture. Anyone seeing the benchmark results will realize that.

The writers of ScienceMark are hardware enthusiasts, just like you folks -- we love seeing how our code performs on different architectures, and if an architecture is strong in our code, we have no problem shouting it from the rooftops.

[Fred_Pohl]

Nah, clearing inventory for socket F which will be released soon.

Great job with the chip OP, too bad you can't OC it. Personlly I'd like to see how well it does on air and phase, dice scores are always fun to gawk at too.

All those saying AMD=dead, you keep forgetting they'll have thier .65 process in volume in Q1 2007, that is only about 3 months or so from the launch date of Conroe. I don't think they'll be able to take the lead back on performance but I'm pretty sure reduce it significantly, at the same price point vs. Intel counterparts. Things will get interesting again when K8L comes out in late 2007 for AMD, til' then they'll have to compete on price I think, which should help keep Intel prices from getting to high. Either which way it sounds like I'll be able to get a good chip for a reasonable price, go competition go!!

Conroe is slated to launch in Q3-06 and most indications are that it will be early in Q3. AMD says that they will introduce 65nm desktop cpus sometime in Q1-07 but do you really think they can ramp them to well over 3GHz that quickly to minimize Conroe's huge performance advantage? The 3 months you speak of looks more like 6 months to me and even then, we're not talking about any sort of performance parity but only closing the gap a little bit.

As for pricing, AMD will have to cut their prices in half or more just to acheive some sort of price/performance parity with Conroe, if currently released prices hold true. This time around it isn't Intel we have to worry about overcharging for their cpus, it's AMD. $1236 for a FX62 is a joke when Intel will be offering a faster E6600 for $316.

[Fred_Pohl]

According to the preview from Anandtech 2.66GHz does beat FX60@2.8GHz in games and other applications.

Indeed it did. By 15-30%!

[Fred_Pohl]

Do you guys think



it is time



to buy stock in Intel??? (looks like the stock price has gone down a lot)

what do you think?

IMHO it's a great time to sell overvalued AMD stock before it plummets again and buy undervalued Intel stock. AMD is not a bad investment but it's stock price is artificially high right now and Intel's is artificially low. AMD is struggling to borrow money to buy 65nm equipment for fab 36 and facing a huge price/performance deficit for the next year while Intel is as financially sound as ever with a vastly superior cpu architecture about to launch in a few months. Although AMD's stock value has larger peaks and valleys, making it a better potential short-term investment, it's current price is about as high as it will ever go. Intel doesn't swing from $15-40/share as AMD does but $20/share is a bargain since it will almost certainly hit close to $30/share again within a year.

[Your_Boss]

Hi, First post.

Anyway, I've been a long time viewer of about 4 years now but just now registering, bad me.

Conroe looks good so far of what I've seen. My guess is it will get 17 sec in 1M SPi if clocked to about 3GHz. Just my guess. I will look out for AM2 as well when it arrives.

[zabomb4163]

problem with intel stock right now is that analyst predict lower profit margins this year (bad for profits)

[RimRam]

Tell me about it.
And some have closed the book as thu conroe/AM2 have come and gone, and perfectly clear that "conroe is smoking AMD", "AMD is bankrupt because of conroe", and we conroe users just have to buy Intel stocks and AMD users have to sell AMD stocks. CLOWNS
STOP TRASHING

[JoeBar]

hiho everybody

well, I've done some google-search and found this about the Albatron PX975X. sadly


http://www.techpowerup.com/reviews/Cebit2006/Day3

but hey, there are another Mobos up to come from Albatron which have the digital clock generator and the metal mosfets as well ^^
gimme more power harhar

I really would like to check this mobo out. Too bad that it's not available yet...

[AgentVX]

I've managed to get one. Should have it this weekend...

Godspeed. Hopefully we'll finally see some overclocking results from Conroe

[TarTheDark]

I've brought this and another link to this thread. Question is whether it is sure or only rumours.

Well I guess we'll find out soon enough, won't we?

As for the overclocking potential of Conroe, it should be very good. At 2.4 GHz it only needs passive cooling and it's a 65 nm part.

[automagic]

redpriest quote "What I am perplexed about is why Conroe bombs on the encryption code"

Does this mean Conroe will not be as strong in video encoding as it is in other areas?

Thanks

[uOpt]

redpriest quote "What I am perplexed about is why Conroe bombs on the encryption code"

Does this mean Conroe will not be as strong in video encoding as it is in other areas?

You cannot draw that conclusion for high-quality video encoding.

Most one-key encryption schemes and low-quality video encoding have a tendency not to care about the L2 CPU cache at all because they rip through so many data blocks (or frames squares) that memory bandwidth counts a lot. Things are either in the L1 cache or they go right through the L2 cache. Conroe is probably weaker here because of the lack of internal memory controller.

However, high-quality video encoding spends so much time on each frame that memory access doesn't play a big role. Same if you don't only encrypt the data but you do something with the data before encryption. Only "pure" encryption is subject to this.

The benchmarks in my signature illustrate that, too. It is clear that memory bandwidth and latency is most worthwhile with low-quality video encoding. But for high-quality video encoding even a stupid Celeron with no memory bandwidth and small L2 cache can be very fast.

[White Wizard]

In regards to SM2 and Video Coding:

SM2, as redpriest stated, is a work of enthusiasm that begain when I was in college with version 1.0. Version 2.0 has Alex and I working collectively upon it. As Alex, redpriest, stated, we purchase hardware and software out of our own pockets and do not recieve support from Intel or AMD. Case in point, I just purchased for 400$ the Intel MKL library for hoisting the DGEMM and SGEMM testing currently done with my own code out of SM2 and leverage Intel's MKL and AMD's ACML. Soon, in a new release, SM2 will test FFTs, Linear Alegebra, be multi-threaded and have support for new bnechmarks from a vectorizing compiler.

In regards to media and encryption. Encryption is very tightly bound by dependencies, SHA1, SHA256, RC4, etc. The greatest optimizers achieve almost 1.5-nearly 3 IPC presently upon their algorithms. Like Alex, I had expected more from Conroe upon this benchmark with it's 4 way issue of ALU ops. In media, SIMD vector instructions are heavily utilized, in MP3 popularly with MMX while video codecs use SSE2 and soon MNI. These vector instructions will run +2x better upon the "core" arch. What is witnessed in the BLAS benchmark results of SM2 will also hold true in Media apps with SIMD vector optimizations.

my 2 cents for what its worth..

[dinos22]

In regards to SM2 and Video Coding:

SM2, as redpriest stated, is a work of enthusiasm that begain when I was in college with version 1.0. Version 2.0 has Alex and I working collectively upon it. As Alex, redpriest, stated, we purchase hardware and software out of our own pockets and do not recieve support from Intel or AMD. Case in point, I just purchased for 400$ the Intel MKL library for hoisting the DGEMM and SGEMM testing currently done with my own code out of SM2 and leverage Intel's MKL and AMD's ACML. Soon, in a new release, SM2 will test FFTs, Linear Alegebra, be multi-threaded and have support for new bnechmarks from a vectorizing compiler.

In regards to media and encryption. Encryption is very tightly bound by dependencies, SHA1, SHA256, RC4, etc. The greatest optimizers achieve almost 1.5-nearly 3 IPC presently upon their algorithms. Like Alex, I had expected more from Conroe upon this benchmark with it's 4 way issue of ALU ops. In media, SIMD vector instructions are heavily utilized, in MP3 popularly with MMX while video codecs use SSE2 and soon MNI. These vector instructions will run +2x better upon the "core" arch. What is witnessed in the BLAS benchmark results of SM2 will also hold true in Media apps with SIMD vector optimizations.

my 2 cents for what its worth..

english pls

[zakelwe]

Thanks for the SM clarifications guys ..appreciated.

Regards

Andy