Sorry for the double-post on edit guys, see above/my next post, interesting info. on many grounds imo... in a hurry while home @ lunch posting it...

:)

APK

---

http://torry.net/authorsmore.php?id=1781

"The object's hull is made of SOLID neutronium: A single StarShip cannot combat it!" quote Mr. Spock, Star Trek original series, episode title: "The Doomsday Machine"

I have an Intel 2.8Mhz Prescott on ASUS board running at 3.0 Ghz, and an AMD 64 X2 4200+_in HP box and board. The Intel has 512Mb ram while the AMD has 1Gb. The Intel shows higher MIPS in both FPU and Integer ops and consistently increases it's lead over the AMD by about 200 to 250 "Total Credit" units per day. However the "Results" column shows usually shows around 55 to 60 for the AMD and only 48 to 52 units for the Intel. What gives here?

There is a larger backlog of Pending credit for the Intel machine, about 33% more than for AMD at any given point of time.

It's not considered good form to hijack someone else's thread by changing the topic/subject. Your own new thread is the place to ask this question.

Respectfully,

Jim

---

Without love, breath is just a clock ... ticking.
Equilibrium

I have an Intel 2.8Mhz Prescott on ASUS board running at 3.0 Ghz, and an AMD 64 X2 4200+_in HP box and board. The Intel has 512Mb ram while the AMD has 1Gb. The Intel shows higher MIPS in both FPU and Integer ops and consistently increases it's lead over the AMD by about 200 to 250 "Total Credit" units per day. However the "Results" column shows usually shows around 55 to 60 for the AMD and only 48 to 52 units for the Intel. What gives here?

There is a larger backlog of Pending credit for the Intel machine, about 33% more than for AMD at any given point of time.

---

When we finally figure it all out, all the rules will change and we can start all over again.

I just want to point out that, these future products aside, Intel is already in the lead when it comes to SETI@home... just for using larger caches. AMD is clearly in the lead when it comes to gaming and server performance, but relatively cheap 1600 MHz Pentium M laptops outperform my 3500+ Athlon 64 (Venice, SSE3, 1mb cache) desktop on SETI@home simply because they have a 2mb cache.

Cache size doesn't much matter with gaming machines or for servers. But while gamers are better off with Athlon 64 (at least until these new Intel parts come out), crunchers are better off with Pentium D (or even Pentium M laptops!). It's all about the cache size, really.

With regard to Intel's new Conroe part that this thread is about, it depends on which version. As noted, any Conroe is a jump by leaps and bounds over Pentium D for gaming. But the low-end Conroes with 2mb cache won't be much better than the Pentium D for crunching. What is really exciting here is that the 4mb cache versions of Conroe will be inconceivably awesome for SETI@home and, while I am proud my AMD 64 chip does 1.3 work units per hour, the 8mb cache "Extreme Edition" Conroes will probably churn out 10 or more work units per hour!!!

In regards to SETI@home, Cache is King. So count me in for switching to from Athlon 64 to Conroe when the 8mb Extreme Editions are semi-affordable (yes, that may take quite a while). Of course, this all depends on whether or not the new Enhanced SETI@home still puts such a huge emphasis on cache size versus system bus speed. If it becomes more in line with other BOINC projects, AMD may catch back up in SETI@home even as it falls a little behind in gaming.

In conclusion, server administrators and web hosting companies should stick with AMD (Opteron) versus Conroe because system bus and HyperTransport is what matters here. Gamers should switch to Conroe, but only if Conroe is affordable... it probably won't be, because knowing Intel's business model as we all do, the future Athlon 64s will continue to be sold at an attractive price point verus Conroe. So many gamers will probably want to upgrade with Athlon 64s.

The only group that should switch from AMD to Intel (before or after Conroe) with absolute certainty are SETI@home crunching enthusiasts, and we would already be better off with Pentium D.

---

Yikes guys, have a beer or three. Chill.

The facts seem to be that Intel has finally done something right for the first time in several years. Ever since I've been building computers (only 5 years now admittedly) AMD has been the clear winner in gaming and general value given per dollar invested. That's not contestable. Memory bandwidth has been Intel's only advantage throughout this time. So finally they come out with a kick ass part that seems to be able to do it all. That only means that AMD will be spurred on to make some big innovations to recapture the value position again. That's been their strong point all along. They're experts at the game of catch up. And again, their talent for it has made them an equal contender instead of an underdog.

All this means is that we, the enthusiasts, will benefit in the long run. I can't wait to see what the climate will be come end of summer. I may even build a Conroe system next.

More than anything else, it just makes sense to wait and see what things will be like around January and February next year. By then there will be enough boards out to drive prices down and incentivise the board makers to sell also by virtue of offering richly featured boards. AMD will certainly have announced a competitor by then, adding to the pot.

Remember that we won't even see the first trickle of really expensive 1st issues until Q3 - provided Intel stays on schedule. Investing in Conroe before Jan.-Feb. will mean that you will be spending money in that 3-4 month period where you're on the cutting edge, but unless you're wealthy, but will regret the extra $300-500 you spent on a new, less fully featured board and the new cpu. Add another $150-200 to that if you don't have DDR2 memory yet. It's a very well established pattern, one we've seen for many years.

I'm excited about all this. It's WAY too early to panic or cry victory. These are battles. The war will never end until one of them buys the other.

Another beer please. It's 7:30 AM already.

Jim

---

Without love, breath is just a clock ... ticking.
Equilibrium

Intel is not even officially out of the gate with their new product.

So?
When pre-production samples are running as well as that to dismiss it as irrelevant would be rather foolish.

Oh, are you one of those who count your winnings when the bookie hands you fresh warm manure from the stables?

I don't gamble.
Nor do i add 2+2 & come up with answers other than 4. Take a few deep breaths & everything will be allright.

---

Grant
Darwin NT

Intel have not only caught up, they've taken the lead.

What do you mean "they've taken the lead."?

Intel is not even officially out of the gate with their new product.
How many thousands have you seen in users hands?

The AMD Athlon 64 FX they are comparing it with is out there running real work.

Oh, are you one of those who count your winnings when the bookie hands you fresh warm manure from the stables?

The background theme to all this is that AMD have a much more effective design at present and that Intel are working hard to try to "catch up". The question is what AMD will have in their new designs by the time Intel have caught up with the present.

Intel have not only caught up, they've taken the lead.
See here for the numbers.

---

Grant
Darwin NT

More material in the never ending comparisons between Intel CPUs and AMD CPUs for the argument of which might be 'better'...

See: Intel Developers' Forum Spring 2006: Will Intel's Core Architecture Close the Technology Gap?

That article gives an interesting overview of past and future Intel CPU trends and how development has been shaped by the competition with AMD.

One brief quote:

If you are aware of recent processor history, Intel's new strategy is a very logical reaction. Pentium 4 and Pentium D processors draw more power than their AMD counterparts the Athlon 64 and Athlon 64 X2. This translates into higher cooling requirements and a higher energy bill. As soon as systems run 24/7, this difference easily adds up to $100 a year in North America or considerably more in countries with higher energy costs. Also think of large corporations with 100s or 1000s of systems: The energy cost difference can be a competitive disadvantage.

[...]

Performance = Frequency * Instructions per clock cycle

At last year's fall IDF, the announcement was to beat the competitor both in absolute performance and in performance per Watt. Intel changed that statement even more and talked about "satisfaction per Watt," which brings all the current processor features such as 64-bit capability and virtualization technology into the equation. In the end, it all comes down to delivering as much value as possible per clock cycle without involving ridiculous thermals.

And Intel goes a step further: It does not matter how long the processor pipeline actually is, it does not matter whether the memory controller is integrated or not, and it does not matter what clock speed the device actually is running at. All that matters is to deliver maximum performance at minimum power requirements. It does sound good, doesn't it? What we want now is proof.

For my analysis of the above:

I agree with the first paragraph there. The following paragraphs however have some subtle Marketing distortions to distract you away from the full truth.

A more honest equation is:
Performance = Frequency * Instructions per clock cycle * Instructions Usefulness

Also note the new Marketing term of "satisfaction per Watt". Very sweet marketing to warm over the real technical requirements.

And note: "It does not matter ... the processor ... actually is, ... All that matters is to deliver maximum performance at minimum power requirements." Again, this draws you away from the very good design decisions that are already in AMD's CPUs to instead look blindly towards the 'end result'. Note that CPUs are a very careful balance of design features. Benchmarks are so notoriously unusable precisely because only specific features are measured. Real-world usage tests the balance of all the features that are exercised. Hence, "maximum performance" can be whatever you want it to be... regardless of real-world reality! The CPU internals and the performance balance achieved are always important. Unless that is, the performace is vastly more than you need for your particular task...

The background theme to all this is that AMD have a much more effective design at present and that Intel are working hard to try to "catch up". The question is what AMD will have in their new designs by the time Intel have caught up with the present.

Enjoy!

Happy crunchin',
Martin

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)