PROOF - Conroe Was Not Faithful

[qurious69ss]

You know I look at it like this. An F1 engine from say ferrari can be concidered architecturaly superior to that of hondas' but if that honda wins the race then whose engine is better? Now I'm not trying to compare speed to hz, I'm comparing speed to higher benchmarks...

 

[Viperabyss]

i understand. but intel has a very good reputation of giving false benchmarks before the release of the product. setting conroe aside, prescott's benchmark test was modified that gave a tremendous edge for intel. later it shows that prescott was not the same as intel claimed.

and this time, it rose many suspicion because intel did not show a lot of improvement on conroe/merom/woodcrest's architectur. however, intel bragged that the processor had at least 20% better of performance over any current high end processors. i seriously doubt switching from 90nm to 65nm would really improve its architecture, since conroe still uses a FSB or DIB = =...

just like the racecar theory you just gave. ferrari has a better engine than honda. but honda's team manager boast that honda's new engine will outperform ferrari's engine by at least 10 seconds, while honda's engine still has the similar architecture as honda's old engine. wouldn't that raise a lot of questions and doubts?

 

[qurious69ss]

True, they have failed before......I think that if they fail again even their PR deparment won't be able to get help them.

 

[Action_Man]

and haven't come up with a real dual core architectures

They've had real dual core since yonah.

their chips have very limited potential

And you're basing this off what now?

Merom, Conroe, and Woodcrest, are new chips designed with power consumption with mind, but their architectures are not as radical as what amd did with athlon xp and a64.

Conroe is way more different then the P4 then the A64 was to the AXP.

i seriously doubt switching from 90nm to 65nm would really improve its architecture

Oh definately they just used the newer process because they wanted something to do.

Stop trolling.

 

[bront]

Personaly, untill a chipset is close to release, I'd view any benchmarks as speculitive.

If they realy did want these benchmarks done, why didn't they offer a few sample chips to several reviewers instead of running demos themselves or including only one reviewer?

I'm not saying that Conroe won't be great, but there is no need to get so excited over results that are speculitive and on an unfinished product. It could be a hand picked chip, or they may have to do a revision that will change the performance, or they could have fixed it. Doesn't matter, it's still 6 months away.

 

[Action_Man]

why didn't they offer a few sample chips to several reviewers instead of running demos themselves or including only one reviewer?

New steppings, limited supply atm, etc. Hexus also got to use it too and probably some other people too.

It could be a hand picked chip

Very unlikely since Woodcrest clocks at 3.0.

 

[AmdMELTDOWN]

"Set Cool 'n' Quiet Default to Disabled

- With Cool & Quiet enabled, AMD processors will throttle in order to save power and bring their thermal load down. This means the processor could be running as low as 800MHz in certain programs – no matter what the program is. In theory Cool & Quiet is supposed to throttle up to maximum in games but this is not always the case. No enthusiast PC goes out with Cool & Quiet enabled unless it’s a fanless machine or media center.

lol, Amd chip throttles now? so if a normal pc user buys an Amd system, they'll be getting 800Mhz worth???

 

[bront]

lol, Amd chip throttles now? so if a normal pc user buys an Amd system, they'll be getting 800Mhz worth???

Wow, You're still around too?

The idea is to reduce power draw when the CPU isn't fully in use. Apparently this could have done better, but it probably does save heat and power for most home users, so it's perfectly reasonable. But for a benchmark, I'd probably disable all that.

 

[MU_Engineer]

Okay, four things:

1. The chips go down to 1000 MHz, not 800MHz.
2. You can control the chip frequency exactly like you can for a laptop chip via ACPI settings. By default, it is set to a dynamic-scaling mode where the chip will throttle up and down with the load. In Linux, set the cpufreqd governor to "performance" or in Windows, set it to "Home and Office Desk" to get full CPU speed at all times.
3. Or you can just disable it in BIOS.
4. Intel chips have EIST, which is pretty much the same.

Processor throttling is not a new thing at all. I am frankly surprised to have seen it migrate to desktops as late as it did.

 

[RichPLS]

Just show him your AMD cpu and he'll take back everything *holds up his venice*

Oh, I didn't know he wanted my validation permit? [/slaps forehead]

 

[RichPLS]

Holds up Opteron 175 and my uber overclocked Athlon XP 1800+...
I/we come in peace 😉

 

[luminaris]

Now who's the dumbass?

"Set Cool 'n' Quiet Default to Disabled

Turns out, the follow up story just shredded all of your evidence. The Cool n Quiet was disabled already and BIOS issue turns out, was not an issue at all.

This thread has got to make you feel ashamed. Insulting people while spamming an article from some computer company CEO that IMO, hardly exists. That article alone gives him a discredit to his name.

http://www.dailytech.com/article.aspx?newsid=1185

 

[Viperabyss]

Now who's the dumbass?

"Set Cool 'n' Quiet Default to Disabled

Turns out, the follow up story just shredded all of your evidence. The Cool n Quiet was disabled already and BIOS issue turns out, was not an issue at all.

This thread has got to make you feel ashamed. Insulting people while spamming an article from some computer company CEO that IMO, hardly exists. That article alone gives him a discredit to his name.

http://www.dailytech.com/article.aspx?newsid=1185
well.. as it turns out, the follow up shows conroe has a 20% increase in FEAR rather than the previous 41 % increase. so intel's brag about their "40% increase in performance and power efficiency" is basically a myth. so anyway.. where is the fact for 40% increase in power efficiency?

 

[RichPLS]

That is when it is released based on the optimizations Intel is currently working on.
Remember, this is the mid-line 2.66GHz model, they will release a 3GHz version at launch, and possibly even a 3.33GHz one, both of which will be considerably faster than the benchmarks you are referring to.

 

[luminaris]

Well, the test they performed did not measure power efficiency so, I can't answer that. The TCP of conroe is considerably lower then the P4s but right off the top of my head, I don't know what those numbers are.